[bugfix]: select architecture-specific Blackwell kernel targets

- Detect GB200 and GB300 as 10.0a and 10.3a
- Build future GB200 CI images with 10.0a
- Add regression tests for detection, cache keys, and build arguments

.agents/skills/add-model-01-prep/scripts/clone_reference_repo.py

@@ -10,9 +10,7 @@ from pathlib import Path
 
 
 def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(
-        description="Clone a reference repo for FastVideo parity tests."
-    )
+    parser = argparse.ArgumentParser(description="Clone a reference repo for FastVideo parity tests.")
     parser.add_argument("repo_url", help="Official reference repository URL")
     parser.add_argument("target_dir", help="Directory to clone into")
     parser.add_argument("--branch", help="Branch or tag to clone")
@@ -62,9 +60,7 @@ def gitignore_entry_for(target: Path) -> str:
     try:
         relative = resolved.relative_to(root)
     except ValueError as exc:
-        raise ValueError(
-            "--update-gitignore requires target_dir to be under the current directory"
-        ) from exc
+        raise ValueError("--update-gitignore requires target_dir to be under the current directory") from exc
 
     text = relative.as_posix().rstrip("/")
     return "/" + text + "/"

.agents/skills/add-model-01-prep/scripts/download_hf_weights.py

@@ -8,14 +8,12 @@ import os
 import sys
 from pathlib import Path
 
-
 HF_TOKEN_ENV_KEYS = ("HF_TOKEN", "HUGGINGFACE_HUB_TOKEN", "HF_API_KEY")
 
 
 def parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser(
-        description="Download a HF model snapshot or selected files into a local directory."
-    )
+        description="Download a HF model snapshot or selected files into a local directory.")
     parser.add_argument("repo_id", help="HF repo id, for example Org/Model")
     parser.add_argument("local_dir", help="Destination directory")
     parser.add_argument("--repo-type", default="model", help="HF repo type (default: model)")

.agents/skills/add-model-01-prep/scripts/inspect_hf_layout.py

@@ -10,7 +10,6 @@ import sys
 from pathlib import Path
 from typing import Any
 
-
 HF_TOKEN_ENV_KEYS = ("HF_TOKEN", "HUGGINGFACE_HUB_TOKEN", "HF_API_KEY")
 RAW_WEIGHT_SUFFIXES = (".safetensors", ".pt", ".pth", ".ckpt", ".bin")
 KNOWN_COMPONENTS = {
@@ -34,8 +33,7 @@ KNOWN_COMPONENTS = {
 
 def parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser(
-        description="Classify a HF repo or local directory as Diffusers, raw, custom, or unknown."
-    )
+        description="Classify a HF repo or local directory as Diffusers, raw, custom, or unknown.")
     parser.add_argument("source", help="HF repo id or local weights directory")
     parser.add_argument("--repo-type", default="model", help="HF repo type (default: model)")
     parser.add_argument("--revision", help="HF revision to inspect")
@@ -94,14 +92,12 @@ def load_remote_files(
 ) -> list[str]:
     from huggingface_hub import list_repo_files
 
-    return sorted(
-        list_repo_files(
-            repo_id,
-            repo_type=repo_type,
-            revision=revision,
-            token=token,
-        )
-    )
+    return sorted(list_repo_files(
+        repo_id,
+        repo_type=repo_type,
+        revision=revision,
+        token=token,
+    ))
 
 
 def load_remote_model_index(
@@ -215,24 +211,24 @@ def build_result(args: argparse.Namespace) -> dict[str, Any]:
         "components_seen": components,
         "file_count": len(files),
         "file_scan_truncated": truncated,
-        "files_sample": files[: args.sample_limit],
+        "files_sample": files[:args.sample_limit],
     }
 
 
 def print_human(result: dict[str, Any]) -> None:
     for key in (
-        "source",
-        "source_kind",
-        "repo_type",
-        "revision",
-        "token_env",
-        "source_layout",
-        "needs_conversion",
-        "model_index_class",
-        "model_index_diffusers_version",
-        "model_index_error",
-        "file_count",
-        "file_scan_truncated",
+            "source",
+            "source_kind",
+            "repo_type",
+            "revision",
+            "token_env",
+            "source_layout",
+            "needs_conversion",
+            "model_index_class",
+            "model_index_diffusers_version",
+            "model_index_error",
+            "file_count",
+            "file_scan_truncated",
     ):
         value = result.get(key)
         if value is not None:

.agents/skills/add-model-02-parity/templates/component_parity_test.py

@@ -18,7 +18,6 @@ import pytest
 import torch
 from torch.testing import assert_close
 
-
 os.environ.setdefault("MASTER_ADDR", "localhost")
 os.environ.setdefault("MASTER_PORT", "29519")
 os.environ.setdefault("DISABLE_SP", "1")
@@ -35,15 +34,10 @@ FASTVIDEO_CONFIG_CLASS = "<FastVideoConfig>"  # TODO.
 FASTVIDEO_MODEL_MODULE = "fastvideo.models.<bucket>.<module>"  # TODO.
 FASTVIDEO_MODEL_CLASS = "<FastVideoModel>"  # TODO.
 
-OFFICIAL_REF_DIR = Path(
-    os.getenv("<FAMILY_UPPER>_OFFICIAL_REF_DIR", REPO_ROOT / "<ReferenceDir>")
-)
-LOCAL_WEIGHTS_DIR = Path(
-    os.getenv("<FAMILY_UPPER>_LOCAL_WEIGHTS_DIR", REPO_ROOT / "official_weights" / FAMILY)
-)
-CONVERTED_WEIGHTS_DIR = Path(
-    os.getenv("<FAMILY_UPPER>_CONVERTED_WEIGHTS_DIR", REPO_ROOT / "converted_weights" / FAMILY)
-)
+OFFICIAL_REF_DIR = Path(os.getenv("<FAMILY_UPPER>_OFFICIAL_REF_DIR", REPO_ROOT / "<ReferenceDir>"))
+LOCAL_WEIGHTS_DIR = Path(os.getenv("<FAMILY_UPPER>_LOCAL_WEIGHTS_DIR", REPO_ROOT / "official_weights" / FAMILY))
+CONVERTED_WEIGHTS_DIR = Path(os.getenv("<FAMILY_UPPER>_CONVERTED_WEIGHTS_DIR",
+                                       REPO_ROOT / "converted_weights" / FAMILY))
 
 
 def _resolve_hf_token() -> str | None:
@@ -99,18 +93,14 @@ def _load_official_model(device: torch.device, dtype: torch.dtype) -> torch.nn.M
     model = OfficialClass()  # TODO: pass official config kwargs.
     state_dict = {}  # TODO: load official state dict from LOCAL_WEIGHTS_DIR.
     missing, unexpected = model.load_state_dict(state_dict, strict=True)
-    assert not missing and not unexpected, (
-        f"official load mismatch missing={missing[:5]} unexpected={unexpected[:5]}"
-    )
+    assert not missing and not unexpected, (f"official load mismatch missing={missing[:5]} unexpected={unexpected[:5]}")
     return model.to(device=device, dtype=dtype).eval()
 
 
 def _load_fastvideo_model(device: torch.device, dtype: torch.dtype) -> torch.nn.Module:
     """Load the FastVideo component with the same tensor content."""
     if not CONVERTED_WEIGHTS_DIR.exists() and not LOCAL_WEIGHTS_DIR.exists():
-        pytest.skip(
-            f"No FastVideo loadable weights: {CONVERTED_WEIGHTS_DIR} or {LOCAL_WEIGHTS_DIR}"
-        )
+        pytest.skip(f"No FastVideo loadable weights: {CONVERTED_WEIGHTS_DIR} or {LOCAL_WEIGHTS_DIR}")
 
     # TODO: replace with the bucket-specific FastVideo config/class/loader.
     # DiT examples:
@@ -127,8 +117,7 @@ def _load_fastvideo_model(device: torch.device, dtype: torch.dtype) -> torch.nn.
     state_dict = {}  # TODO: load converted or directly mapped state dict.
     missing, unexpected = model.load_state_dict(state_dict, strict=True)
     assert not missing and not unexpected, (
-        f"FastVideo load mismatch missing={missing[:5]} unexpected={unexpected[:5]}"
-    )
+        f"FastVideo load mismatch missing={missing[:5]} unexpected={unexpected[:5]}")
     return model.to(device=device, dtype=dtype).eval()
 
 
@@ -187,11 +176,9 @@ def test_component_parity():
 
     assert official_out.shape == fastvideo_out.shape
     diff = (official_out - fastvideo_out).abs()
-    print(
-        f"official abs_mean={official_out.abs().mean().item():.6f} "
-        f"fastvideo abs_mean={fastvideo_out.abs().mean().item():.6f} "
-        f"diff_max={diff.max().item():.6f} diff_mean={diff.mean().item():.6f}"
-    )
+    print(f"official abs_mean={official_out.abs().mean().item():.6f} "
+          f"fastvideo abs_mean={fastvideo_out.abs().mean().item():.6f} "
+          f"diff_max={diff.max().item():.6f} diff_mean={diff.mean().item():.6f}")
 
     # TODO: pick tolerance by scope:
     # - single block / same kernel: 1e-4

.agents/skills/add-model-07-conversion/templates/family_to_diffusers.py

@@ -27,7 +27,6 @@ try:
 except ImportError:  # pragma: no cover - optional local conversion dependency
     snapshot_download = None
 
-
 # TODO: fill with authoritative component prefixes for monolithic checkpoints.
 # Example: {"model.model.": "transformer", "pretransform.model.": "vae"}
 COMPONENT_PREFIXES: dict[str, str] = {}
@@ -47,10 +46,7 @@ SKIP_PATTERNS: tuple[str, ...] = ()
 
 
 def _hf_token() -> str | None:
-    return (
-        os.environ.get("HF_TOKEN") or os.environ.get("HUGGINGFACE_HUB_TOKEN")
-        or os.environ.get("HF_API_KEY")
-    )
+    return (os.environ.get("HF_TOKEN") or os.environ.get("HUGGINGFACE_HUB_TOKEN") or os.environ.get("HF_API_KEY"))
 
 
 def resolve_src(src: str, revision: str | None) -> Path:
@@ -95,11 +91,10 @@ def apply_mapping(key: str) -> str | None:
     return key
 
 
-def split_monolithic(
-    state: dict[str, torch.Tensor],
-) -> dict[str, OrderedDict[str, torch.Tensor]]:
+def split_monolithic(state: dict[str, torch.Tensor], ) -> dict[str, OrderedDict[str, torch.Tensor]]:
     components: dict[str, OrderedDict[str, torch.Tensor]] = {
-        name: OrderedDict() for name in set(COMPONENT_PREFIXES.values())
+        name: OrderedDict()
+        for name in set(COMPONENT_PREFIXES.values())
     }
     intentionally_skipped: list[str] = []
     unowned: list[str] = []
@@ -117,10 +112,8 @@ def split_monolithic(
             unowned.append(key)
     if unowned:
         sample = ", ".join(unowned[:10])
-        raise ValueError(
-            f"Unowned monolithic keys: {len(unowned)}. "
-            f"Add COMPONENT_PREFIXES or SKIP_PATTERNS entries. Sample: {sample}"
-        )
+        raise ValueError(f"Unowned monolithic keys: {len(unowned)}. "
+                         f"Add COMPONENT_PREFIXES or SKIP_PATTERNS entries. Sample: {sample}")
     if intentionally_skipped:
         print(f"Intentionally skipped {len(intentionally_skipped)} keys")
     return {name: weights for name, weights in components.items() if weights}
@@ -143,8 +136,12 @@ def build_component_configs(_src_dir: Path) -> dict[str, dict[str, Any]]:
     # TODO: emit config content accepted by FastVideo loaders. Most components use
     # config.json; schedulers use scheduler_config.json.
     return {
-        "transformer": {"_class_name": "<FastVideoTransformerClass>"},
-        "vae": {"_class_name": "<FastVideoVAEClass>"},
+        "transformer": {
+            "_class_name": "<FastVideoTransformerClass>"
+        },
+        "vae": {
+            "_class_name": "<FastVideoVAEClass>"
+        },
     }
 
 
@@ -177,19 +174,13 @@ def build_model_index(
     }
     if revision:
         index["_fastvideo_converted_revision"] = revision
-    return {
-        key: value
-        for key, value in index.items()
-        if key.startswith("_") or key in available_components
-    }
+    return {key: value for key, value in index.items() if key.startswith("_") or key in available_components}
 
 
 def validate_component_configs(configs: dict[str, dict[str, Any]]) -> None:
     # TODO: instantiate each FastVideo config and call update_model_arch(...) or
     # update_model_config(...) with this JSON so unknown emitted keys fail here.
-    placeholder_configs = [
-        name for name, config in configs.items() if "<" in json.dumps(config)
-    ]
+    placeholder_configs = [name for name, config in configs.items() if "<" in json.dumps(config)]
     if placeholder_configs:
         raise ValueError(f"Replace config placeholders for: {placeholder_configs}")
 
@@ -201,9 +192,7 @@ def verify_conversion(
     del dst_dir, components
     # TODO: load each emitted stateful component through its production loader and
     # assert strict load, or document exact allowed missing/unexpected keys.
-    raise NotImplementedError(
-        "Implement production config validation and strict-load checks"
-    )
+    raise NotImplementedError("Implement production config validation and strict-load checks")
 
 
 def write_component(
@@ -216,9 +205,7 @@ def write_component(
     if component_dir.exists() and any(component_dir.iterdir()):
         shutil.rmtree(component_dir)
     component_dir.mkdir(parents=True, exist_ok=True)
-    save_file(
-        dict(state), str(component_dir / "diffusion_pytorch_model.safetensors")
-    )
+    save_file(dict(state), str(component_dir / "diffusion_pytorch_model.safetensors"))
     if config is not None:
         config_path = component_dir / config_filename(name)
         with config_path.open("w", encoding="utf-8") as f:
@@ -261,9 +248,7 @@ def convert(
 
     if layout in {"monolithic", "raw_official"}:
         # TODO: replace model.safetensors with the official monolithic file name.
-        components = split_monolithic(
-            load_checkpoint(default_monolithic_checkpoint(src_path))
-        )
+        components = split_monolithic(load_checkpoint(default_monolithic_checkpoint(src_path)))
     elif layout in {"separate_components", "mixed"}:
         if not src_path.is_dir():
             raise ValueError(f"{layout} layout requires a source directory: {src_path}")
@@ -271,9 +256,7 @@ def convert(
     else:
         raise ValueError(f"Unsupported template layout: {layout}")
 
-    copied = (
-        copy_passthrough(src_path, dst_dir) if src_path.is_dir() else []
-    )
+    copied = (copy_passthrough(src_path, dst_dir) if src_path.is_dir() else [])
     configs = build_component_configs(src_path if src_path.is_dir() else src_path.parent)
     validate_component_configs(configs)
     for name, state in components.items():
@@ -289,9 +272,7 @@ def convert(
 
 def main() -> None:
     parser = argparse.ArgumentParser(description=__doc__)
-    parser.add_argument(
-        "--src", required=True, help="HF repo id, local dir, or checkpoint path"
-    )
+    parser.add_argument("--src", required=True, help="HF repo id, local dir, or checkpoint path")
     parser.add_argument("--revision", help="HF branch, tag, or commit for repo sources")
     parser.add_argument(
         "--dst",

.agents/skills/add-model-08-trace/templates/block_trace_debug.py

@@ -24,8 +24,8 @@ from typing import Any
 
 import torch
 
-FAMILY: str = "<family>"       # e.g. "magi_human", "ltx2", "wan"
-COMPONENT: str = "<component>" # e.g. "dit", "vae", "encoder"
+FAMILY: str = "<family>"  # e.g. "magi_human", "ltx2", "wan"
+COMPONENT: str = "<component>"  # e.g. "dit", "vae", "encoder"
 DRILL_LAYER_ENV: str = "<FAMILY>_DEBUG_DRILL_LAYER"
 HYPOTHESIS_ENV: str = "<FAMILY>_DEBUG_PATCH_<HYPOTHESIS>"
 REL_THRESHOLD: float = 0.005  # 0.5% abs_mean drift flags a block as divergent
@@ -94,6 +94,7 @@ def _attach_block_hooks(
     handles: list[Any] = []
 
     def _hook(name: str):
+
         def fn(_module, _inputs, outputs):
             t = outputs[0] if isinstance(outputs, tuple) else outputs
             if not torch.is_tensor(t):
@@ -101,6 +102,7 @@ def _attach_block_hooks(
             log.append({"side": label, **_stat(name, t)})
             if tensors is not None:
                 tensors[name] = t.detach().float().cpu()
+
         return fn
 
     def _pre_hook(name: str):
@@ -114,6 +116,7 @@ def _attach_block_hooks(
             log.append({"side": label, **_stat(key, t)})
             if tensors is not None:
                 tensors[key] = t.detach().float().cpu()
+
         return fn
 
     # TODO: adapt attribute paths to your model. Remove adapter block if absent.
@@ -131,43 +134,21 @@ def _attach_block_hooks(
             # magi-human uses: attention, mlp.pre_norm, mlp.up_gate_proj,
             # mlp.down_proj (pre+post), mlp, attn_post_norm, mlp_post_norm.
             if hasattr(layer, "attention"):
-                handles.append(
-                    layer.attention.register_forward_hook(_hook(f"{tag}.attention"))
-                )
+                handles.append(layer.attention.register_forward_hook(_hook(f"{tag}.attention")))
             if hasattr(layer, "mlp"):
                 mlp = layer.mlp
                 if hasattr(mlp, "pre_norm"):
-                    handles.append(
-                        mlp.pre_norm.register_forward_hook(_hook(f"{tag}.mlp.pre_norm"))
-                    )
+                    handles.append(mlp.pre_norm.register_forward_hook(_hook(f"{tag}.mlp.pre_norm")))
                 if hasattr(mlp, "up_gate_proj"):
-                    handles.append(
-                        mlp.up_gate_proj.register_forward_hook(
-                            _hook(f"{tag}.mlp.up_gate_proj")
-                        )
-                    )
+                    handles.append(mlp.up_gate_proj.register_forward_hook(_hook(f"{tag}.mlp.up_gate_proj")))
                 if hasattr(mlp, "down_proj"):
-                    handles.append(
-                        mlp.down_proj.register_forward_pre_hook(
-                            _pre_hook(f"{tag}.mlp.down_proj")
-                        )
-                    )
-                    handles.append(
-                        mlp.down_proj.register_forward_hook(_hook(f"{tag}.mlp.down_proj"))
-                    )
+                    handles.append(mlp.down_proj.register_forward_pre_hook(_pre_hook(f"{tag}.mlp.down_proj")))
+                    handles.append(mlp.down_proj.register_forward_hook(_hook(f"{tag}.mlp.down_proj")))
                 handles.append(mlp.register_forward_hook(_hook(f"{tag}.mlp")))
             if hasattr(layer, "attn_post_norm"):
-                handles.append(
-                    layer.attn_post_norm.register_forward_hook(
-                        _hook(f"{tag}.attn_post_norm")
-                    )
-                )
+                handles.append(layer.attn_post_norm.register_forward_hook(_hook(f"{tag}.attn_post_norm")))
             if hasattr(layer, "mlp_post_norm"):
-                handles.append(
-                    layer.mlp_post_norm.register_forward_hook(
-                        _hook(f"{tag}.mlp_post_norm")
-                    )
-                )
+                handles.append(layer.mlp_post_norm.register_forward_hook(_hook(f"{tag}.mlp_post_norm")))
     return handles
 
 
@@ -193,11 +174,9 @@ def _write_log(entries: list[dict], path: Path) -> None:
     path.parent.mkdir(parents=True, exist_ok=True)
     with open(path, "w") as f:
         for e in entries:
-            f.write(
-                f"{e['name']} {e['shape']} "
-                f"{e['abs_mean']:.8f} {e['sum']:.4f} "
-                f"{e['min']:.6f} {e['max']:.6f}\n"
-            )
+            f.write(f"{e['name']} {e['shape']} "
+                    f"{e['abs_mean']:.8f} {e['sum']:.4f} "
+                    f"{e['min']:.6f} {e['max']:.6f}\n")
 
 
 def _sort_key(name: str, drill_layer: int) -> tuple:
@@ -205,9 +184,14 @@ def _sort_key(name: str, drill_layer: int) -> tuple:
         return (0, "")
     if name.startswith(f"L{drill_layer:02d}."):
         sub_order = {
-            "attention": 0, "attn_post_norm": 1, "mlp.pre_norm": 2,
-            "mlp.up_gate_proj": 3, "mlp.down_proj<in>": 4,
-            "mlp.down_proj": 5, "mlp": 6, "mlp_post_norm": 7,
+            "attention": 0,
+            "attn_post_norm": 1,
+            "mlp.pre_norm": 2,
+            "mlp.up_gate_proj": 3,
+            "mlp.down_proj<in>": 4,
+            "mlp.down_proj": 5,
+            "mlp": 6,
+            "mlp_post_norm": 7,
         }.get(name.split(".", 1)[1], 9)
         return (1, f"block[{drill_layer:02d}]", sub_order)
     if name.startswith("block["):
@@ -216,10 +200,8 @@ def _sort_key(name: str, drill_layer: int) -> tuple:
 
 
 def _print_table(by_name: dict[str, dict], drill_layer: int) -> int | None:
-    hdr = (
-        f"{'name':<18} {'up_shape':<22} {'up_absmean':>12} {'fv_absmean':>12} "
-        f"{'absmean_diff':>14} {'rel%':>8} {'up_sum':>14} {'fv_sum':>14} {'sum_diff':>12}"
-    )
+    hdr = (f"{'name':<18} {'up_shape':<22} {'up_absmean':>12} {'fv_absmean':>12} "
+           f"{'absmean_diff':>14} {'rel%':>8} {'up_sum':>14} {'fv_sum':>14} {'sum_diff':>12}")
     print(f"\n{hdr}\n{'-' * len(hdr)}")
     first_div: int | None = None
     for name in sorted(by_name.keys(), key=lambda n: _sort_key(n, drill_layer)):
@@ -235,11 +217,9 @@ def _print_table(by_name: dict[str, dict], drill_layer: int) -> int | None:
             flag = " <<< DIVERGE"
             if first_div is None:
                 first_div = int(name[len("block["):-1])
-        print(
-            f"{name:<18} {str(up['shape']):<22} {up['abs_mean']:>12.6f} "
-            f"{fv['abs_mean']:>12.6f} {am_diff:>14.6f} {am_rel * 100:>7.3f}% "
-            f"{up['sum']:>14.4f} {fv['sum']:>14.4f} {sum_diff:>12.4f}{flag}"
-        )
+        print(f"{name:<18} {str(up['shape']):<22} {up['abs_mean']:>12.6f} "
+              f"{fv['abs_mean']:>12.6f} {am_diff:>14.6f} {am_rel * 100:>7.3f}% "
+              f"{up['sum']:>14.4f} {fv['sum']:>14.4f} {sum_diff:>12.4f}{flag}")
     return first_div
 
 
@@ -255,10 +235,8 @@ def _print_elementwise(up_t: dict[str, torch.Tensor], fv_t: dict[str, torch.Tens
             continue
         diff = (a - b).abs()
         rel = (diff.mean().item() / max(a.abs().mean().item(), 1e-9)) * 100
-        print(
-            f"{name:<30} {str(tuple(a.shape)):<22} "
-            f"{diff.max().item():>12.6f} {diff.mean().item():>12.6f} {rel:>9.4f}%"
-        )
+        print(f"{name:<30} {str(tuple(a.shape)):<22} "
+              f"{diff.max().item():>12.6f} {diff.mean().item():>12.6f} {rel:>9.4f}%")
 
 
 def main() -> None:

.agents/skills/add-model-09-pipeline/templates/pipeline_parity_test.py

@@ -43,12 +43,10 @@ def _add_official_to_path() -> Path:
 
 def _log_tensor_stats(label: str, tensor: torch.Tensor) -> None:
     value = tensor.detach().float()
-    print(
-        f"[{_MODEL_FAMILY} PIPELINE] {label}: shape={tuple(tensor.shape)} "
-        f"dtype={tensor.dtype} device={tensor.device} "
-        f"min={value.min().item():.6f} max={value.max().item():.6f} "
-        f"mean={value.mean().item():.6f} std={value.std().item():.6f}"
-    )
+    print(f"[{_MODEL_FAMILY} PIPELINE] {label}: shape={tuple(tensor.shape)} "
+          f"dtype={tensor.dtype} device={tensor.device} "
+          f"min={value.min().item():.6f} max={value.max().item():.6f} "
+          f"mean={value.mean().item():.6f} std={value.std().item():.6f}")
 
 
 def _extract_tensor(output: Any, key: str) -> torch.Tensor:
@@ -73,10 +71,8 @@ def _run_official_pipeline(
     device: torch.device,
 ) -> Any:
     del official_path, params, device
-    pytest.skip(
-        "TODO: import the official pipeline/factory, load official weights, "
-        "run with params, and return the comparison target."
-    )
+    pytest.skip("TODO: import the official pipeline/factory, load official weights, "
+                "run with params, and return the comparison target.")
 
 
 def _run_fastvideo_pipeline(model_path: Path, params: dict[str, Any]) -> Any:
@@ -146,8 +142,6 @@ def test_todo_model_family_pipeline_official_parity() -> None:
     assert official_tensor.shape == fastvideo_tensor.shape
 
     diff = (official_tensor - fastvideo_tensor).abs()
-    print(
-        f"diff max={diff.max().item():.6f} "
-        f"mean={diff.mean().item():.6f} median={diff.median().item():.6f}"
-    )
+    print(f"diff max={diff.max().item():.6f} "
+          f"mean={diff.mean().item():.6f} median={diff.median().item():.6f}")
     assert_close(fastvideo_tensor, official_tensor, atol=1e-2, rtol=1e-2)

.agents/skills/ci-runner/SKILL.md

@@ -0,0 +1,99 @@
+---
+name: ci-runner
+description: Work on FastVideo's Slurm-only, change-aware GPU CI lanes, static Buildkite graph, trusted ci-runner policy, lane scripts, and GB200 validation.
+---
+
+# Slinky Slurm CI lanes
+
+FastVideo's `ci-runner` Buildkite queue is the control plane for all active
+GPU CI. A private host-owned dispatcher leases GPUs from the Slinky Slurm tray
+and runs the immutable PR SHA inside an isolated Enroot container. Buildkite
+pipeline upload and Slurm submission occur on the login plane; every test
+payload executes on Slurm compute.
+
+The files under `fastvideo/tests/modal/` and `.buildkite/scripts/pr_test.sh`
+are dormant rollback code. Never add an active Buildkite or slash-command
+route to them. `pr_test.sh` must continue to reject Buildkite invocations.
+
+The private operator bundle is deliberately outside this repository because
+it contains site paths and credentials. See
+`docs/contributing/ci_architecture.md`; this skill covers the repository half
+and the coordination contract with that bundle.
+
+## Invariants
+
+- `.buildkite/pipeline.yml` contains exactly one static step for every active
+  GPU lane. Each step pins a unique key and label, a 90-minute timeout, the
+  trusted `/opt/fastvideo-ci-runner/run-ci` command (`run-unit` is the one
+  compatibility wrapper), step-level internal `TEST_TYPE`, and
+  `queue: "ci-runner"`.
+- Active CI contains no `pr_test.sh` command, Modal invocation, default queue,
+  Buildkite plugin, `soft_fail`, or job-controlled artifact glob.
+- The six Fastcheck lanes use `:microscope:` labels. Full-Suite-only lanes use
+  `:test_tube:` or `:bar_chart:` so direct reruns update the right aggregate.
+- SSIM and vanilla training request all four GPUs. Keep both in the
+  `fastvideo/slinky/whole-tray` Buildkite concurrency group with a limit of one
+  so the second job does not consume an agent or command timeout while waiting
+  for the same tray.
+- `/test full` schedules all twenty lanes. `/merge`, `ready`, and new pushes to
+  ready PRs use the trusted base-branch planner in
+  `.github/scripts/plan_merge_ci.py`: automatic Fastcheck remains the universal
+  six-lane baseline, and the merge build adds only path-relevant integration
+  lanes. Unknown source/build paths fail closed to all fourteen additive lanes.
+  The trusted uploader still normalizes and validates the complete static graph
+  before Buildkite evaluates its plan conditions.
+- Focused merge builds may pass allowlisted golden-gate and SSIM test basenames.
+  The private host validates the lane plan and basenames before staging them,
+  and the in-container scripts validate them again. Direct `/test ssim`,
+  explicit `/test full`, and the weekly main-branch schedule run the complete
+  SSIM matrix.
+- The trusted uploader serves exactly three entry pipelines:
+  `pr-fastcheck` for automatic PR builds, `ci` for slash-command/ready-label
+  API builds, and `fastvideo-performance-lane` for the weekly schedule. Keep
+  incoming GitHub webhook processing disabled on `ci` so it cannot duplicate
+  `pr-fastcheck` on every PR update.
+- Test payloads live in `.buildkite/scripts/unit_test.sh` or executable
+  `.buildkite/scripts/lanes/<lane>.sh`. Backend policy (GPU count, extras,
+  secrets, kernel build, artifacts) stays in the agent-owned lane table.
+- Tests must preserve an inherited `MASTER_PORT`. Packed containers share the
+  tray network namespace, so the private runner assigns a distinct port range
+  per GPU lease and the SSIM scheduler assigns task offsets within its range.
+- The ARM64 runner image includes the pinned FA4 CuTe overlay validated on
+  GB200. Keep SSIM at `FASTVIDEO_FA4=1` because its references were seeded with
+  FA4; keep lanes with FA2 baselines at `FASTVIDEO_FA4=0`. A runner image change
+  must revalidate both the FA4 import and an actual GB200 forward kernel.
+- `fastvideo/tests/ssim/ci_runner.py` is the active four-GPU SSIM scheduler.
+  New SSIM files are discovered through `REQUIRED_GPUS` and
+  `*_MODEL_TO_PARAMS`; do not wire them through the dormant Modal scheduler.
+- The host policy fail-closes unknown tuples. A repository-side lane change is
+  inert until the operator updates the private lane table and uploader policy
+  in the same rollout.
+
+## Adding or changing a lane
+
+1. Read the closest `AGENTS.md` and the domain-specific testing guide.
+2. Add or update the executable lane payload under `.buildkite/scripts/`.
+   Keep it deterministic and free of host-specific paths or credential fetches.
+3. Add the static pipeline step and canonical `/test <name>` mapping. Keep the
+   `<name>-ci` alias only when compatibility requires it.
+4. Add its source/test path ownership to `.github/scripts/plan_merge_ci.py`.
+   Prefer the narrowest correctness-preserving lane set; leave unknown paths
+   fail-closed. Extend `fastvideo/tests/contract/test_ci_test_collection.py`,
+   `test_merge_ci_plan.py`, and focused CPU-only scheduler/policy tests.
+5. Coordinate the private lane row: GPU count (1-4), wall time, script, scope
+   pairs, step key, command, HF cache/token, tracking mode, extras, attention
+   backend policy, kernel policy, and artifact relay. Active training lanes
+   keep W&B offline and do not stage a W&B credential.
+6. Update the trusted pipeline-uploader schema. A mismatch must reject the
+   pipeline rather than silently skip a lane.
+7. Run `pre-commit run --files <changed paths>`, the planner's representative
+   diff matrix, contract tests, private driver tests, and a real GB200 canary.
+   Multi-GPU, hardware-reference, training, performance, and SSIM changes need
+   their own target-hardware evidence.
+
+## Rollback
+
+Rollback the Slurm routing/configuration change or pause the `ci-runner` queue.
+Do not silently reactivate Modal. A manual Modal experiment requires the
+explicit local opt-in documented in `ci_architecture.md`; returning it to
+production CI needs a separate reviewed decision.

.agents/skills/env-var-conventions/SKILL.md

@@ -0,0 +1,76 @@
+---
+name: env-var-conventions
+description: Add, read, rename, or remove an environment variable in FastVideo, or change the environment-variable policy. Use before touching fastvideo/envs.py, os.environ, os.getenv, or monkeypatch.setenv in fastvideo/, and when fastvideo/tests/contract/test_env_policy.py fails.
+---
+
+# Environment Variable Conventions
+
+## Purpose
+
+FastVideo registers its environment variables as typed fields in
+`fastvideo/envs.py`. The policy that governs them is
+`docs/contributing/env_vars.md`, and the contract test
+`fastvideo/tests/contract/test_env_policy.py` enforces the policy in the unit
+CI lane. This skill routes an environment-variable change through that policy.
+The policy doc is the single source of the rules; read it instead of relying
+on a summary here.
+
+## Prerequisites
+
+- Read `docs/contributing/env_vars.md` in full.
+- Decide whether the setting belongs in an environment variable or an argument
+  (rule 5 in the policy doc). Settings that users change per deployment are
+  arguments; add them through `fastvideo/fastvideo_args.py` instead.
+
+## Inputs
+
+| Parameter  | Required | Description                                                    |
+| ---------- | -------- | -------------------------------------------------------------- |
+| `change`   | Yes      | Add, read, rename, or remove a variable, or change the policy. |
+| `variable` | Yes      | The variable name, with the `FASTVIDEO_` prefix.               |
+
+## Steps
+
+1. **Declare or edit the variable in `fastvideo/envs.py`.**
+   - Pick the field type and category that the policy doc lists.
+   - Write a description that states what the variable does and its units.
+   - To rename, keep the old name in `deprecated_names`. To remove, add the
+     name to `DEPRECATED_VARIABLES`. Update the uses in `examples/`,
+     `scripts/`, `docs/`, `apps/`, and the tests.
+2. **Read the variable with `envs.NAME.get()` inside a function.**
+   - In tests, change the value with `envs.NAME.override(value)`.
+   - Do not call `os.environ`, `os.getenv`, or `monkeypatch.setenv` for a
+     FastVideo variable.
+   - To set a variable that another tool reads, call `envs.set_external`,
+     `envs.setdefault_external`, or `envs.unset_external`.
+3. **Regenerate the table in the policy doc.**
+   - Run `python fastvideo/tests/contract/test_env_policy.py`.
+4. **Run the contract test.**
+   - Run `pytest fastvideo/tests/contract/test_env_policy.py`.
+   - When the test reports a fixed known violation, delete or lower its entry
+     in `KNOWN_VIOLATIONS`. Never add an entry to `KNOWN_VIOLATIONS`.
+5. **When the policy itself changes, update the policy doc and the contract
+   test in the same pull request.**
+   - The rules in `docs/contributing/env_vars.md`, the checks and allowlist in
+     `fastvideo/tests/contract/test_env_policy.py`, and this skill must agree.
+
+## Outputs
+
+- A registry entry in `fastvideo/envs.py` and call sites that use
+  `envs.NAME.get()`.
+- A regenerated table in `docs/contributing/env_vars.md`.
+- A passing `fastvideo/tests/contract/test_env_policy.py`.
+
+## Example Usage
+
+```
+Add a FASTVIDEO_DEBUG_MY_STAGE switch that logs MyStage inputs.
+```
+
+## References
+
+- `docs/contributing/env_vars.md`: the policy, the field types, and the
+  violation kinds that the contract test reports.
+- `fastvideo/envs.py`: the registry.
+- `fastvideo/tests/contract/test_env_policy.py`: the contract test,
+  `EXTERNAL_ALLOWLIST`, and `KNOWN_VIOLATIONS`.

.agents/skills/reseed-performance-baseline/SKILL.md

@@ -1,20 +1,23 @@
 ---
 name: reseed-performance-baseline
-description: Re-seed the HF performance-tracking baseline for an intentional runtime, dependency, or environment-caused benchmark shift using one or more reviewed normalized performance JSONs. Use when performance CI fails because metrics such as latency, throughput, component time, or peak memory changed for an accepted reason and the rolling median baseline in FastVideo/performance-tracking must be advanced from a consistent batch of reviewed source results. The workflow backs up existing history under /tmp, validates all source JSONs for the same (model_id, gpu_type), rejects internally inconsistent source batches, uploads one success=true reseed record per accepted source JSON, and offers to clean local temp state after a successful upload.
+description: Re-seed the HF performance-tracking baseline for an intentional runtime, dependency, environment-caused benchmark shift, or reviewed v2 calibration using one or more reviewed normalized performance JSONs. Use when performance CI fails because metrics such as latency, throughput, component time, or peak memory changed for an accepted reason and the rolling median baseline in FastVideo/performance-tracking must be advanced, or when a new v2 exact comparable identity needs its first approved baseline. The workflow backs up existing history under /tmp, validates all source JSONs for the same legacy (model_id, gpu_type) target or the same v2 exact identity, rejects internally inconsistent source batches, uploads one success=true baseline record per accepted source JSON, and offers to clean local temp state after a successful upload.
 ---
 
 # Re-seed Performance Baseline
 
 ## Purpose
 
-Replace or advance the rolling performance baseline for a single
-`(model_id, gpu_type)` pair in the HF dataset
-`FastVideo/performance-tracking`.
+Replace or advance the rolling performance baseline in the HF dataset
+`FastVideo/performance-tracking`. Legacy targets are scoped by
+`(model_id, gpu_type)`. V2 targets are scoped by exact comparable identity:
+`workload_id`, `variant_id`, `benchmark_version`, `hardware_profile_id`,
+`software_profile_id`, and `recipe_fingerprint`.
 
-Performance comparison uses the median of up to the last 5 successful records
-for the same model and GPU. Failed records are useful audit history, but they
-do not move the future baseline because `compare_baseline.py` loads records
-with `successful_only=True`.
+Performance comparison uses the median of up to the last 5 successful,
+baseline-eligible records for the same target. Failed or calibration-only
+records are useful audit history, but they do not move the future baseline
+because `compare_baseline.py` loads records with `successful_only=True` and
+`baseline_eligible_only=True`.
 
 This skill now reseeds from a reviewed batch of one or more source performance
 JSONs. It uploads one new `success=true` record per accepted source JSON; it
@@ -22,11 +25,13 @@ does not blindly replicate one measurement into 3 or 5 records. The effective
 reseed size is therefore dynamic and equals the number of provided, validated,
 internally consistent source JSONs.
 
-If the operator provides fewer than 3 records, call out that the last-5 rolling
-median may not move immediately. If the operator provides 3 consistent shifted
-records, the rolling median usually moves immediately. If the operator provides
-5 consistent shifted records, the last-5 window is effectively reset to the new
-runtime profile.
+For baseline shifts with existing history, if the operator provides fewer than
+3 records, call out that the last-5 rolling median may not move immediately. If
+the operator provides 3 consistent shifted records, the rolling median usually
+moves immediately. If the operator provides 5 consistent shifted records, the
+last-5 window is effectively reset to the new runtime profile. For the first
+approved v2 baseline of a new exact identity, one reviewed calibration seed is
+enough for the next comparable run to leave `CALIBRATION_NEEDED`.
 
 These records are intentional operator-approved baseline resets, not ordinary
 independent main-branch persistence. Mark them clearly with provenance fields
@@ -66,8 +71,8 @@ approval, then upload reviewed accepted baseline records.
 
 | Parameter | Required | Description |
 |-----------|----------|-------------|
-| `model_id` | Yes | Benchmark id, e.g. `wan-t2v-1.3b-2gpu`. This maps to the HF subdirectory after `sanitize(model_id)`. |
-| `gpu_type` | Yes | Exact GPU device string from the performance record, e.g. the L40S device name emitted by CI. Baselines are GPU-specific. |
+| `model_id` | Legacy required; v2 inferred | Benchmark id, e.g. `wan-t2v-1.3b-2gpu`. This maps to the HF subdirectory after `sanitize(model_id)`. For v2 records, use the `model_id` from each source artifact only as the upload directory; comparison is by exact identity. |
+| `gpu_type` | Legacy required; v2 inferred | Exact GPU device string from the performance record, e.g. the L40S device name emitted by CI. V2 hardware matching uses `hardware_profile_id`; preserve `gpu_type` as display metadata. |
 | `source_results` | Yes | One or more local paths or Buildkite artifact URLs for accepted shifted performance JSONs. Prefer normalized `normalized_perf_*.json` artifacts emitted by `compare_baseline.py`. Accept `source_result` as an alias only for a single JSON. |
 | `max_intra_batch_regression` | No | Maximum allowed regression of any source JSON against the source batch median. Default: `0.05` (5%). |
 | `intent_rationale` | Yes | One-line explanation for why the baseline shift is legitimate. This is written into provenance and should be reused in the PR. |
@@ -78,10 +83,14 @@ Hardcoded defaults:
   supported by the code, but use the default unless the user explicitly asks).
 - Local sync root: `/tmp/perf-tracking` (`PERFORMANCE_TRACKING_ROOT` override
   is supported).
+- Prepared-record staging root: `/tmp/performance_reseed_prepared`
+  (`PERFORMANCE_RESEED_STAGING_ROOT` override is supported). Keep it separate
+  and non-nested from the sync root.
 - Backup root: `/tmp/performance_reseed_backup`.
 - Download scratch root for source artifact URLs: `/tmp/performance_reseed_source`.
-- Baseline window: last 5 `success=true` records for the same
-  `(model_id, gpu_type)`.
+- Baseline window: last 5 `success=true`, `baseline_eligible=true` records
+  for the same legacy `(model_id, gpu_type)` target or the same v2 exact
+  comparable identity.
 - Reseed count: dynamic. Upload exactly one accepted seed record per validated
   source JSON.
 
@@ -115,12 +124,24 @@ with open(source_result, encoding="utf-8") as f:
     record = json.load(f)
 ```
 
-Stop if any normalized record's `model_id` or `gpu_type` does not match the
-requested `model_id` and `gpu_type`.
+Classify the source batch before continuing:
+
+- **Legacy source records** have no v2 exact identity fields. Stop if any
+  normalized record's `model_id` or `gpu_type` does not match the requested
+  `model_id` and `gpu_type`.
+- **V2 source records** have exact identity fields. Stop unless every source
+  record has all six comparable identity fields and they are identical across
+  the batch: `workload_id`, `variant_id`, `benchmark_version`,
+  `hardware_profile_id`, `software_profile_id`, and `recipe_fingerprint`.
+  Do not fall back to legacy `(model_id, gpu_type)` matching for v2 records.
 
 The source records may have `success: false` when they came from failed
 rolling baseline comparisons. That is expected; only the reviewed reseed
 records become new `success: true` baseline records after explicit approval.
+For a first v2 baseline seed, the source records must instead be successful
+scheduled-main full-suite `CALIBRATION_NEEDED` normalized artifacts. Reject PR,
+local, direct-run, non-main-branch, or non-full-suite calibration artifacts as
+seed sources.
 
 Sort validated source records by their original `timestamp` ascending before
 preparing the seed records. If a source timestamp is missing or unparsable,
@@ -194,7 +215,7 @@ export HF_REPO_ID="${HF_REPO_ID:-FastVideo/performance-tracking}"
 python -c 'from fastvideo.performance.hf_store import sync_from_hf; import os; sync_from_hf(os.environ["PERFORMANCE_TRACKING_ROOT"], strict=True)'
 ```
 
-Then back up only the sanitized model directory under `/tmp`:
+For legacy records, back up the sanitized model directory under `/tmp`:
 
 ```bash
 SHORT_COMMIT=$(git rev-parse --short=12 HEAD)
@@ -209,6 +230,16 @@ mkdir -p "$BACKUP_DIR"
 cp -R "${PERFORMANCE_TRACKING_ROOT}/${MODEL_SAFE}" "$BACKUP_DIR/" 2>/dev/null || true
 ```
 
+For v2 records, back up the full local tracking root after sync. Exact identity
+lookup scans across model directories, so a benchmark rename may have relevant
+history outside the current source artifact's `model_id` directory:
+
+```bash
+BACKUP_DIR="/tmp/performance_reseed_backup/${TIMESTAMP}_${SHORT_COMMIT}_v2_exact_identity"
+mkdir -p "$BACKUP_DIR"
+cp -R "${PERFORMANCE_TRACKING_ROOT}" "$BACKUP_DIR/tracking-root"
+```
+
 Write provenance next to the backup:
 
 ```bash
@@ -231,7 +262,9 @@ first baseline seed. Continue, but report that baseline history was empty.
 
 ### 3. Compute old baseline and candidate shift
 
-Load the last 5 successful records for the target:
+Load the last 5 successful baseline records for the target.
+
+For legacy targets:
 
 ```python
 from fastvideo.performance.hf_store import load_records_for_model
@@ -242,6 +275,28 @@ records = load_records_for_model(
     "<gpu_type>",
     last_n=5,
     successful_only=True,
+    baseline_eligible_only=True,
+)
+```
+
+For v2 exact-identity targets:
+
+```python
+from fastvideo.performance.hf_store import load_records_for_identity
+
+records = load_records_for_identity(
+    "/tmp/perf-tracking",
+    {
+        "workload_id": "<workload_id>",
+        "variant_id": "<variant_id>",
+        "benchmark_version": "<benchmark_version>",
+        "hardware_profile_id": "<hardware_profile_id>",
+        "software_profile_id": "<software_profile_id>",
+        "recipe_fingerprint": "<recipe_fingerprint>",
+    },
+    last_n=5,
+    successful_only=True,
+    baseline_eligible_only=True,
 )
 ```
 
@@ -257,7 +312,8 @@ medians after appending the proposed seed records, and source batch spread for:
 
 Also print how many successful old records exist. Make clear:
 
-- 1 seed record usually does not move a last-5 median by itself.
+- 1 seed record usually does not move an existing last-5 median by itself, but
+  it is enough to establish the first v2 baseline for a new exact identity.
 - 3 consistent seed records usually move the last-5 median immediately.
 - 5 consistent seed records effectively reset the last-5 window.
 - The records are intentional approved baseline resets and must be labeled
@@ -267,10 +323,10 @@ Also print how many successful old records exist. Make clear:
 
 Require an explicit confirmation phrase before preparing the upload:
 
-> About to RE-SEED performance baseline for `<model_id>` on `<gpu_type>`.
+> About to RE-SEED performance baseline for `<target description>`.
 > This will upload `<N>` new `success=true` records to
-> `FastVideo/performance-tracking/<sanitize(model_id)>/`, one per accepted
-> source JSON.
+> `FastVideo/performance-tracking/<sanitize(model_id)>/` or the source
+> artifact's v2 model directory, one per accepted source JSON.
 >
 > Reason: `<intent_rationale>`
 > Source results: `<source_results>`
@@ -288,25 +344,63 @@ Do not continue unless the user types exactly `confirm performance reseed`.
 
 ### 5. Create the accepted seed records
 
-Create one seed record from each normalized source result. Do not copy the
+Create one seed record from each normalized source result.
+
+For first v2 baseline seeds, use the scoped utility. It validates exact
+identity, requires successful scheduled-main full-suite `CALIBRATION_NEEDED`
+source artifacts, preserves the normalized v2 identity and metadata fields,
+and writes seed records with `success=true`, `baseline_eligible=true`, and
+`comparison_status=PASS`:
+
+```bash
+python fastvideo/tests/performance/seed_baseline.py \
+  --source-result <normalized_perf_1.json> \
+  --source-result <normalized_perf_2.json> \
+  --intent-rationale "<intent_rationale>" \
+  --max-intra-batch-regression 0.05 \
+  --tracking-root "${PERFORMANCE_TRACKING_ROOT}" \
+  --staging-root "${PERFORMANCE_RESEED_STAGING_ROOT:-/tmp/performance_reseed_prepared}"
+```
+
+The utility is prepare-only and intentionally has no upload option. Upload the
+scoped records only after the separate confirmation in step 6.
+
+The utility validates against an isolated fresh HF snapshot and leaves
+`PERFORMANCE_TRACKING_ROOT` untouched; that argument only proves the staging
+root is separate from the operator's tracking mirror. Before writing, it stops
+if the exact identity already has a successful baseline-eligible record or if
+the workload/variant/version already trusts another recipe. It atomically
+reserves the exact identity and writes a digest-protected upload manifest bound
+to the current HF endpoint, repository id, and repository type. Keep the
+prepared records, manifest, source files, and reservation unchanged until the
+operation is uploaded or explicitly cleaned up.
+
+If the prepared seed records look correct, upload only those scoped records in
+step 7. Do not rerun the utility with a different source list after approval.
+
+For legacy reseeds or accepted v2 baseline shifts from regression artifacts,
+create one seed record from each normalized source result. Do not copy the
 source JSON wholesale.
 
 Infer the baseline field allowlist from all existing HF records for the target
-`(model_id, gpu_type)` after syncing, including both `success=true` and
-`success=false` records. Use the union of non-provenance keys present in those
-target records, preserving only fields that also exist in the normalized
-source record or are explicitly set by the reseed workflow. Always include
-`model_id`, `timestamp`, and `success` because the upload path and baseline
-loader depend on them. Always set `timestamp` to a fresh reseed timestamp and
-`success` to `true`. Do not include unrelated source-only fields that are
-absent from existing HF records.
+after syncing, including both `success=true` and `success=false` records. For
+legacy targets the target is `(model_id, gpu_type)`. For v2 baseline-shift
+reseeds the target is the exact comparable identity. Use the union of
+non-provenance keys present in those target records, preserving only fields
+that also exist in the normalized source record or are explicitly set by the
+reseed workflow. Always include `model_id`, `timestamp`, `success`,
+`baseline_eligible`, and `comparison_status` because the upload path and
+baseline loader depend on them. Always set `timestamp` to a fresh reseed
+timestamp, `success` to `true`, `baseline_eligible` to `true`, and
+`comparison_status` to `PASS`. Do not include unrelated source-only fields
+that are absent from existing HF records.
 
 Exclude existing provenance or operator metadata from the inferred baseline
 field allowlist. At minimum, exclude keys prefixed with `baseline_reseed` and
 any fields known to be local-only audit metadata.
 
-If there are no previous HF records for the target model/GPU, fall back to this
-default baseline field list:
+If there are no previous HF records for the target, fall back to this default
+baseline field list:
 
 - `model_id`
 - `timestamp`
@@ -319,6 +413,22 @@ default baseline field list:
 - `dit_time_s`
 - `vae_decode_time_s`
 - `success`
+- `baseline_eligible`
+- `comparison_status`
+
+For v2 baseline-shift reseeds with no previous HF records for the exact
+identity, also preserve:
+
+- `workload_id`
+- `variant_id`
+- `benchmark_version`
+- `recipe_fingerprint`
+- `hardware_profile_id`
+- `software_profile_id`
+- `recipe`
+- `hardware_profile`
+- `software_profile`
+- `software_comparison_profile`
 
 Do not upload extra fields from the source artifact.
 
@@ -334,6 +444,22 @@ Optional provenance fields are allowed and useful:
 - `baseline_reseed_operator`
 - `baseline_reseed_max_intra_batch_regression`
 
+The v2 calibration seed utility writes analogous first-seed provenance:
+
+- `baseline_seed: true`
+- `baseline_seed_reason`
+- `baseline_seed_source_result`
+- `baseline_seed_source_status`
+- `baseline_seed_source_timestamp`
+- `baseline_seed_source_success`
+- `baseline_seed_source_run_source`
+- `baseline_seed_source_branch`
+- `baseline_seed_source_test_scope`
+- `baseline_seed_source_pr_number`
+- `baseline_seed_batch_size`
+- `baseline_seed_batch_index`
+- `baseline_seed_operator`
+
 Use a fresh reseed timestamp for each seed record, not the original source
 result timestamp. This is required because
 `load_records_for_model(..., last_n=5)` keeps the last records after loading
@@ -356,7 +482,8 @@ Prefer uploading new accepted seed records so failed history remains visible.
 Print:
 
 - Backup directory path under `/tmp`.
-- Prepared local record paths under `PERFORMANCE_TRACKING_ROOT`.
+- Prepared local record paths under `PERFORMANCE_RESEED_STAGING_ROOT`.
+- Prepared upload-manifest path under the identity reservation.
 - HF paths that will receive the new records.
 - Old rolling medians.
 - Source batch medians, source batch spread, reseed count, and candidate
@@ -368,22 +495,36 @@ prepared records plus backup on disk.
 
 ### 7. Upload only the scoped records
 
-Use the shared storage helper so the path and repo type match CI:
+For a first v2 calibration seed, use the manifest uploader after the user
+replies exactly `upload`:
 
-```python
-from fastvideo.performance.hf_store import upload_record
-
-upload_record("<local_record_path>", record, strict=True)
+```bash
+python -c 'from fastvideo.tests.performance.seed_baseline import upload_prepared_seed_manifest; print(upload_prepared_seed_manifest("<prepared_manifest>"))'
 ```
 
-Run it once per prepared record. Each upload goes to:
+The uploader verifies the source and prepared-record digests, pins and scans
+the current HF revision, rechecks exact-identity and recipe-cohort conflicts,
+and writes the entire batch in one commit whose `parent_commit` must still be
+current. A concurrent Hub update makes the commit fail. Do not retry
+automatically: preserve staging, refresh/review remote state, and request a new
+explicit `upload` after the conflict is understood. Each record goes to:
 
 ```text
 FastVideo/performance-tracking/<sanitize(model_id)>/<record_filename>.json
 ```
 
-Never bulk upload the whole tracking root. Never modify another model's
-directory in the same operation.
+Never call `upload_record()` once per first-seed record: that can partially
+land the batch and has no compare-and-swap guard.
+
+For a legacy reseed or an accepted v2 baseline shift, the first-seed manifest
+validator does not apply because an eligible baseline already exists. Upload
+only the individually reviewed records prepared in step 5 with the shared
+`upload_record(local_path, record, strict=True)` helper. Stop on the first
+failure and report exactly which records reached HF; do not silently rerun or
+replicate the remainder.
+
+Never bulk upload the tracking or staging root, and never modify another
+model's directory in the same operation.
 
 ### 8. Report outcome and offer cleanup
 
@@ -405,9 +546,14 @@ distinguish an accepted baseline shift from a hidden regression.
 After the upload is verified, ask whether the user wants to clear temporary
 local state. Explain what each directory is for:
 
-- `PERFORMANCE_TRACKING_ROOT`, usually `/tmp/perf-tracking`: local synced
-  mirror of `FastVideo/performance-tracking` plus the prepared local seed
-  records used for scoped upload.
+- `PERFORMANCE_TRACKING_ROOT`, usually `/tmp/perf-tracking`: read-only local
+  synced mirror used for operator review and reporting. First-v2 preparation
+  independently proves remote state from a fresh temporary HF snapshot.
+- `PERFORMANCE_RESEED_STAGING_ROOT`, usually
+  `/tmp/performance_reseed_prepared`: prepared local seed records used for the
+  scoped upload, plus the identity reservation and digest manifest. Keeping
+  this separate prevents aborted preparations from appearing in later
+  baseline reads.
 - `/tmp/performance_reseed_backup/<...>`: local backup of the target model's
   pre-reseed HF history plus `PROVENANCE.txt`, kept so a bad reseed can be
   audited or corrected.
@@ -417,14 +563,18 @@ local state. Explain what each directory is for:
 Ask:
 
 > Reseed succeeded. Do you want me to delete the local temp tracking mirror,
-> source downloads, and reseed backup under `/tmp`? These files are local
-> safety/audit artifacts only; HF already has the uploaded records.
+> this reseed's prepared staging records, source downloads, and reseed backup
+> under `/tmp`? These files are local safety/audit artifacts only; HF already
+> has the uploaded records.
 >
 > Reply `cleanup reseed temp` to delete them, anything else to keep them.
 
 Do not delete anything unless the user replies exactly
 `cleanup reseed temp`. If cleanup is requested, remove only the specific
-directories created for this reseed. Never remove unrelated `/tmp` contents.
+directories and prepared record paths created for this reseed. Do not remove
+the shared staging root when it contains other records. Remove this operation's
+identity reservation only with its prepared records and manifest, and never
+remove unrelated `/tmp` contents.
 
 ## Failure modes and handling
 
@@ -436,19 +586,34 @@ directories created for this reseed. Never remove unrelated `/tmp` contents.
   against the source batch median by more than `max_intra_batch_regression`.
   Ask for cleaner sources or a reviewed explanation before continuing.
 - **Too few source records to move the median.** Continue only after making
-  clear that one or two records may not immediately move the last-5 median.
+  clear that one or two records may not immediately move an existing last-5
+  median. This warning does not block a first v2 calibration seed for an exact
+  identity with no eligible baseline yet.
 - **The source results are noisy or suspicious.** Stop. Reseeding amplifies
   those measurements into the baseline, so they must be reviewed first.
 - **HF sync fails.** Stop for destructive reseeds. A stale or empty sync can
   make the old baseline look missing.
+- **The exact v2 identity already has an eligible baseline.** Stop. The
+  `CALIBRATION_NEEDED` artifact is stale; use the reviewed baseline-shift path
+  instead of the first-seed utility.
+- **The workload/variant/version trusts another recipe.** Stop. The source is
+  stale relative to the current recipe cohort and must not bypass
+  `RECIPE_MISMATCH` by creating a second trusted recipe.
+- **The staging root already has a prepared seed for the exact identity.**
+  Stop and reuse, upload, or explicitly clean that preparation. Do not prepare
+  another copy of the same measurement.
+- **The conditional Hub commit loses its parent race.** Stop without retrying.
+  Keep the preparation, refresh and review the new remote state, then request
+  a new explicit `upload` only if the seed is still valid.
 - **Candidate still violates fixed thresholds.** Report that this skill only
   handles the rolling HF baseline; update benchmark JSON thresholds in code
   review if maintainers accept the new absolute limit.
 - **The user aborts at either confirmation.** Leave the backup and prepared
   records on disk. Nothing should be uploaded.
-- **The user declines cleanup.** Keep `/tmp/perf-tracking`, the source
-  download directory if any, and `/tmp/performance_reseed_backup/<...>` in
-  place for audit/debugging.
+- **The user declines cleanup.** Keep `/tmp/perf-tracking`, the prepared seed
+  records under `/tmp/performance_reseed_prepared`, the source download
+  directory if any, and `/tmp/performance_reseed_backup/<...>` in place for
+  audit/debugging.
 - **A bad seed was uploaded.** Use the backup and HF history to identify the
   uploaded file, then remove or supersede it with an explicitly reviewed
   corrective record. Do not silently rewrite unrelated history.
@@ -459,8 +624,9 @@ directories created for this reseed. Never remove unrelated `/tmp` contents.
   intentional baseline replacement.
 - `fastvideo/tests/performance/compare_baseline.py` — normalization, rolling
   median comparison, and persistence rules.
-- `fastvideo/performance/hf_store.py` — HF sync, record loading,
-  `sanitize()`, and `upload_record()`.
+- `fastvideo/performance/hf_store.py` — HF sync and record loading helpers.
+- `fastvideo/tests/performance/seed_baseline.py` — first-seed preparation,
+  staging reservation, manifest validation, and conditional batch upload.
 - `fastvideo/tests/performance/test_inference_performance.py` — source result
   JSON schema.
 - `.buildkite/performance-benchmarks/tests/*.json` — fixed absolute benchmark
@@ -473,3 +639,4 @@ directories created for this reseed. Never remove unrelated `/tmp` contents.
 | 2026-05-03 | Initial version. Sister workflow to `reseed-ssim-references`, scoped to one performance `(model_id, gpu_type)` baseline seed with backup, confirmation, provenance, and `success=true` upload. |
 | 2026-05-03 | Previous policy: replicate one approved shifted source result into 3 success records by default, or 5 only when explicitly requested. Add provenance marker for replicated-source reseeds. Superseded by the 2026-05-08 dynamic multi-source policy. |
 | 2026-05-08 | Replace fixed 3/5 replication with dynamic multi-source reseeding: upload one seed record per reviewed source JSON, validate intra-batch consistency, move backup/source scratch under `/tmp`, and ask whether to clean temp state after successful upload. |
+| 2026-07-13 | Keep first-v2-seed preparation outside the canonical mirror, reserve staging identities atomically, reject stale or replayed calibration seeds, and upload reviewed manifests with a single parent-guarded Hub commit. |

.agents/skills/reseed-ssim-references/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: reseed-ssim-references
-description: Re-seed HF reference videos for a single existing SSIM test on Modal L40S. Always backs up current refs locally first, regenerates on Modal, pauses for the user to eyeball before-vs-after quality, then overwrites the targeted `<model_id>` subtree on `FastVideo/ssim-reference-videos` with `--force`. Use when an intentional code change (model port fix, attention backend swap, kernel upgrade, hyperparameter change) has invalidated existing refs and they need to be regenerated. Pairs with `seed-ssim-references`, which is for first-time seeding only.
+description: Re-seed HF reference videos for a single existing SSIM test on Modal L40S. Always backs up current refs locally first, regenerates on Modal, pauses for the user to eyeball before-vs-after quality, then overwrites the targeted model subtree on `FastVideo/ssim-reference-videos` with `--force`. Use when an intentional code change (model port fix, attention backend swap, kernel upgrade, hyperparameter change) has invalidated existing refs and they need to be regenerated. Pairs with `seed-ssim-references`, which is for first-time seeding only.
 ---
 
 # Re-seed SSIM Reference Videos
@@ -13,7 +13,7 @@ on HF — the old refs are overwritten — so the skill always:
 
 1. Confirms intent with a one-liner the user has to type.
 2. Downloads the existing refs as a local, timestamped backup.
-3. Regenerates on Modal L40S (same code path that CI uses).
+3. Regenerates through the manual legacy Modal L40S maintenance path.
 4. Pauses for a side-by-side eyeball of backup vs new mp4s.
 5. Uploads with `--force`, scoped to the single `--model-id`.
 6. Reminds the user to keep the backup until the PR lands.
@@ -51,8 +51,9 @@ harder to recover from than failing closed.
 
 Hardcoded:
 
-- Modal GPU: **L40S** (matches CI; re-seeding from another SKU produces refs
-  that L40S CI cannot match).
+- Modal GPU: **L40S**. This is a manual reference-maintenance target, not the
+  active Slurm CI compute path; changing the SKU also changes the historical
+  `L40S_reference_videos` contract.
 - Quality tier: **`default`**. `full_quality` is a separate, deliberate
   operation.
 - HF repo: `FastVideo/ssim-reference-videos` (override via

.agents/skills/seed-ssim-references/SKILL.md

@@ -35,7 +35,8 @@ The skill is run **manually**, once per new test. Before invoking it, the user
 has already sanity-tested the new test locally — it launches `VideoGenerator`
 and writes an artefact without crashing (the missing-reference assertion at
 the end is expected). The skill does not re-test locally; it goes straight
-to Modal L40S (which is what CI uses).
+to the manual legacy Modal L40S reference-maintenance target. Active CI runs
+on the Slinky Slurm cluster and only consumes the resulting references.
 
 ## When to use
 
@@ -61,7 +62,8 @@ Prompt the user for it if they didn't supply it.
 
 Everything else is fixed:
 
-- Modal runner GPU: **L40S** (hardcoded in `fastvideo/tests/modal/ssim_test.py`).
+- Modal maintenance GPU: **L40S** (hardcoded in
+  `fastvideo/tests/modal/ssim_test.py`; this is not the active CI compute path).
 - Device folder: `L40S_reference_videos`.
 - Quality tier: `default` (the tier CI runs). The `full_quality` tier is not
   seeded by this skill.

.buildkite/performance-benchmarks/tests/wan-t2v-1.3b.json

@@ -3,7 +3,7 @@
   "config_schema_version": 2,
   "workload_id": "wan-t2v",
   "variant_id": "1.3b-sp2",
-  "benchmark_version": 2,
+  "benchmark_version": 3,
   "description": "Wan2.1 T2V 1.3B inference performance",
   "model": {
     "model_path": "Wan-AI/Wan2.1-T2V-1.3B-Diffusers",

.buildkite/pipeline.yml

@@ -1,6 +1,9 @@
 env:
   IMAGE_VERSION: "py3.12-latest"
   BUILDKITE_CLEAN_CHECKOUT: true
+  # Slurm workers clone the immutable commit and initialize submodules inside
+  # their isolated container. The Buildkite login-plane checkout is a no-op.
+  BUILDKITE_GIT_SUBMODULES: false
 
 notify:
   - github_commit_status:
@@ -8,528 +11,471 @@ notify:
     if: build.env("TEST_SCOPE") == "fastcheck" || build.env("TEST_SCOPE") == null
   - github_commit_status:
       context: "full-suite-passed"
-    if: build.env("TEST_SCOPE") == "full"
+    if: build.env("TEST_SCOPE") == "full" || build.env("TEST_SCOPE") == "merge"
   - github_commit_status:
       context: "direct-test-completed"
     if: build.env("TEST_SCOPE") == "direct"
+  - github_commit_status:
+      context: "scheduled-ssim-passed"
+    if: build.env("TEST_SCOPE") == "scheduled"
 
+# This is the complete active GPU CI surface. Every command is a trusted host
+# dispatcher, and every test payload executes inside the Slinky Slurm tray.
+# fastvideo/tests/modal remains available only for an explicit manual rollback;
+# no active pipeline or slash-command route invokes it.
 steps:
-  # ============================================================
-  # Direct test: triggered by /test <name> slash command.
-  # Labels match fastcheck/full-suite counterparts so the GitHub
-  # check status overwrites the original failed check.
-  # Only ONE step executes per build (gated by TEST_TYPE).
-  # ============================================================
+  - label: ":microscope: Encoder Tests"
+    key: "encoder"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,encoder,/) ||
+      build.env("TEST_SCOPE") == "fastcheck" ||
+      build.env("TEST_SCOPE") == null ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "encoder" || build.env("TEST_TYPE") == "encoder_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "encoder_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
 
-  # --- Fastcheck-scope direct tests ---
-    - label: ":microscope: Encoder Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "encoder"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":microscope: VAE Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "vae"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":microscope: Transformer Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "transformer"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":microscope: Kernel Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "kernel_tests"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":microscope: Unit Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "unit_test"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":microscope: DreamVerse App Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "dreamverse_app"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
+  - label: ":microscope: VAE Tests"
+    key: "vae"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,vae,/) ||
+      build.env("TEST_SCOPE") == "fastcheck" ||
+      build.env("TEST_SCOPE") == null ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "vae" || build.env("TEST_TYPE") == "vae_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "vae_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
 
-  # --- Full-suite-scope direct tests ---
-    - label: ":bar_chart: SSIM Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "ssim"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-          - exit_status: 1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: LoRA Inference Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "inference_lora"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: LoRA Extraction Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "lora_extraction"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: Training Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "training"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: Distillation DMD Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "distillation_dmd"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: Self-Forcing Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "self_forcing"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: LoRA Training Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "training_lora"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-          - exit_status: 1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: Training Tests VSA"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "training_vsa"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-          - exit_status: 1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: Inference Tests VMoBA"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "inference_vmoba"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: Performance Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "performance"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: API Server Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "api_server"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: Train Framework Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "train_framework"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-    - label: ":test_tube: Eval Metrics Tests"
-      if: build.env("TEST_SCOPE") == "direct" && build.env("TEST_TYPE") == "eval"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
+  - label: ":microscope: Transformer Tests"
+    key: "transformer"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,transformer,/) ||
+      build.env("TEST_SCOPE") == "fastcheck" ||
+      build.env("TEST_SCOPE") == null ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "transformer" || build.env("TEST_TYPE") == "transformer_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "transformer_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
 
-  # ============================================================
-  # Fastcheck: Runs on every PR (~10-15 min parallel)
-  # Core component validation: encoders, VAEs, transformers,
-  # CUDA kernels, and unit tests.
-  # ============================================================
-    - label: "Trigger Fastcheck"
-      if: build.env("TEST_SCOPE") == "fastcheck" || build.env("TEST_SCOPE") == null
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      plugins:
-        - monorepo-diff#v1.4.0:
-            diff: 'git fetch origin "${BUILDKITE_PULL_REQUEST_BASE_BRANCH:-main}" && git diff --name-only "origin/${BUILDKITE_PULL_REQUEST_BASE_BRANCH:-main}...HEAD"'
-            watch:
-            - path:
-                - "fastvideo/models/encoders/**"
-                - "fastvideo/models/loader/**"
-                - "fastvideo/tests/encoders/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 20m .buildkite/scripts/pr_test.sh"
-                label: ":microscope: Encoder Tests"
-                env:
-                  - TEST_TYPE=encoder
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/models/vaes/**"
-                - "fastvideo/models/loader/**"
-                - "fastvideo/tests/vaes/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 20m .buildkite/scripts/pr_test.sh"
-                label: ":microscope: VAE Tests"
-                env:
-                  - TEST_TYPE=vae
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/models/dits/**"
-                - "fastvideo/models/loader/**"
-                - "fastvideo/tests/transformers/**"
-                - "fastvideo/layers/**"
-                - "fastvideo/attention/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: ":microscope: Transformer Tests"
-                env:
-                  - TEST_TYPE=transformer
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo-kernel/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: ":microscope: Kernel Tests"
-                env:
-                  - TEST_TYPE=kernel_tests
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/**"
-                - ".buildkite/**"
-                - ".github/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: ":microscope: Unit Tests"
-                env:
-                  - TEST_TYPE=unit_test
-                agents:
-                  queue: "default"
-            - path:
-                - "apps/dreamverse/**"
-                - "pyproject.toml"
-              config:
-                command: "timeout 30m .buildkite/scripts/pr_test.sh"
-                label: ":microscope: DreamVerse App Tests"
-                env:
-                  - TEST_TYPE=dreamverse_app
-                agents:
-                  queue: "default"
+  - label: ":microscope: Kernel Tests"
+    key: "kernel-tests"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,kernel-tests,/) ||
+      build.env("TEST_SCOPE") == "fastcheck" ||
+      build.env("TEST_SCOPE") == null ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "kernel_tests" || build.env("TEST_TYPE") == "kernel_tests_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "kernel_tests_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
 
-  # ============================================================
-  # Full Suite: Runs when TEST_SCOPE=full
-  # Triggered by adding the 'ready' label (via ci-trigger-full-suite.yml)
-  # or on-demand via /test full slash command.
-  # Includes integration tests, SSIM regression, training pipelines,
-  # and performance benchmarks.
-  # ============================================================
-    - label: "Trigger Full Suite"
-      if: build.env("TEST_SCOPE") == "full"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      plugins:
-        - monorepo-diff#v1.4.0:
-            diff: 'git fetch origin "${BUILDKITE_PULL_REQUEST_BASE_BRANCH:-main}" && git diff --name-only "origin/${BUILDKITE_PULL_REQUEST_BASE_BRANCH:-main}...HEAD"'
-            watch:
-            - path:
-                - "fastvideo/**/*.py"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 90m .buildkite/scripts/pr_test.sh"
-                label: ":bar_chart: SSIM Tests"
-                env:
-                  - TEST_TYPE=ssim
-                retry:
-                  automatic:
-                    - exit_status: 1
-                      limit: 2
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/tests/lora/**"
-                - "fastvideo/models/loader/**"
-                - "fastvideo/tests/transformers/**"
-                - "fastvideo/pipelines/**"
-                - "fastvideo/layers/lora/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 20m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: LoRA Inference Tests"
-                env:
-                  - TEST_TYPE=inference_lora
-                agents:
-                  queue: "default"
-            - path:
-                - "scripts/lora_extraction/**"
-                - "fastvideo/tests/lora_extraction/**"
-                - "fastvideo/models/loader/**"
-                - "fastvideo/training/training_utils.py"
-                - "fastvideo/layers/lora/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 90m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: LoRA Extraction Tests"
-                env:
-                  - TEST_TYPE=lora_extraction
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: Training Tests"
-                env:
-                  - TEST_TYPE=training
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/training/*distillation_pipeline.py"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: Distillation DMD Tests"
-                env:
-                  - TEST_TYPE=distillation_dmd
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/training/*self_forcing_distillation_pipeline.py"
-                - "fastvideo/tests/training/self-forcing/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 30m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: Self-Forcing Tests"
-                env:
-                  - TEST_TYPE=self_forcing
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 25m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: LoRA Training Tests"
-                env:
-                  - TEST_TYPE=training_lora
-                retry:
-                  automatic:
-                    - exit_status: 1
-                      limit: 2
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/**"
-                - "fastvideo-kernel/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: Training Tests VSA"
-                env:
-                  - TEST_TYPE=training_vsa
-                retry:
-                  automatic:
-                    - exit_status: 1
-                      limit: 2
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo-kernel/**"
-                - "fastvideo/attention/backends/vmoba.py"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 15m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: Inference Tests VMoBA"
-                env:
-                  - TEST_TYPE=inference_vmoba
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/models/dits/**"
-                - "fastvideo/pipelines/**"
-                - "fastvideo/attention/**"
-                - "fastvideo/layers/**"
-                - "fastvideo/worker/**"
-                - "fastvideo/entrypoints/**"
-                - "fastvideo/performance/**"
-                - "fastvideo/tests/performance/**"
-                - ".buildkite/performance-benchmarks/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 30m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: Performance Tests"
-                env:
-                  - TEST_TYPE=performance
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/entrypoints/openai/**"
-                - "fastvideo/entrypoints/cli/serve.py"
-                - "fastvideo/tests/entrypoints/test_openai_api_integration.py"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 30m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: API Server Tests"
-                env:
-                  - TEST_TYPE=api_server
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/train/**"
-                - "fastvideo/tests/train/models/**"
-                - "fastvideo/tests/train/fixtures/**"
-                - "fastvideo/models/dits/**"
-                - "fastvideo/models/loader/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 30m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: Train Framework Tests"
-                env:
-                  - TEST_TYPE=train_framework
-                agents:
-                  queue: "default"
-            - path:
-                - "fastvideo/eval/**"
-                - "fastvideo/tests/eval/**"
-                - "pyproject.toml"
-                - "docker/Dockerfile"
-              config:
-                command: "timeout 90m .buildkite/scripts/pr_test.sh"
-                label: ":test_tube: Eval Metrics Tests"
-                env:
-                  - TEST_TYPE=eval
-                agents:
-                  queue: "default"
+  - label: ":microscope: Unit Tests"
+    key: "unit"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,unit,/) ||
+      build.env("TEST_SCOPE") == "fastcheck" ||
+      build.env("TEST_SCOPE") == null ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "unit_test" || build.env("TEST_TYPE") == "unit_test_ci"))
+    command: "/opt/fastvideo-ci-runner/run-unit"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "unit_test_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":microscope: DreamVerse App Tests"
+    key: "dreamverse"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,dreamverse,/) ||
+      build.env("TEST_SCOPE") == "fastcheck" ||
+      build.env("TEST_SCOPE") == null ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "dreamverse_app" || build.env("TEST_TYPE") == "dreamverse_app_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "dreamverse_app_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: Golden-Gate Tests"
+    key: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,golden-gate,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "golden_gate" || build.env("TEST_TYPE") == "golden_gate_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "golden_gate_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":bar_chart: SSIM Tests"
+    key: "ssim"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      build.env("TEST_SCOPE") == "scheduled" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,ssim,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "ssim" || build.env("TEST_TYPE") == "ssim_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    concurrency: 1
+    concurrency_group: "fastvideo/slinky/whole-tray"
+    env:
+      TEST_TYPE: "ssim_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+        - exit_status: 1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: LoRA Inference Tests"
+    key: "lora-inference"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,lora-inference,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "inference_lora" || build.env("TEST_TYPE") == "inference_lora_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "inference_lora_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: LoRA Extraction Tests"
+    key: "lora-extraction"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,lora-extraction,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "lora_extraction" || build.env("TEST_TYPE") == "lora_extraction_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "lora_extraction_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: Training Tests"
+    key: "training"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,training,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "training" || build.env("TEST_TYPE") == "training_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    concurrency: 1
+    concurrency_group: "fastvideo/slinky/whole-tray"
+    env:
+      TEST_TYPE: "training_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: Distillation DMD Tests"
+    key: "distillation"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,distillation,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "distillation_dmd" || build.env("TEST_TYPE") == "distillation_dmd_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "distillation_dmd_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: Self-Forcing Tests"
+    key: "self-forcing"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,self-forcing,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "self_forcing" || build.env("TEST_TYPE") == "self_forcing_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "self_forcing_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: LoRA Training Tests"
+    key: "lora-training"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,lora-training,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "training_lora" || build.env("TEST_TYPE") == "training_lora_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "training_lora_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+        - exit_status: 1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: Training Tests VSA"
+    key: "training-vsa"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,training-vsa,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "training_vsa" || build.env("TEST_TYPE") == "training_vsa_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "training_vsa_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+        - exit_status: 1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: Inference Tests VMoBA"
+    key: "inference-vmoba"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,inference-vmoba,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "inference_vmoba" || build.env("TEST_TYPE") == "inference_vmoba_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "inference_vmoba_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: Performance Tests"
+    key: "performance"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,performance,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "performance" || build.env("TEST_TYPE") == "performance_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "performance_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: API Server Tests"
+    key: "api-server"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,api-server,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "api_server" || build.env("TEST_TYPE") == "api_server_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "api_server_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: Train Framework Tests"
+    key: "train-framework"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,train-framework,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "train_framework" || build.env("TEST_TYPE") == "train_framework_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "train_framework_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"
+
+  - label: ":test_tube: Eval Metrics Tests"
+    key: "eval"
+    depends_on: "golden-gate"
+    if: |
+      build.env("TEST_SCOPE") == "full" ||
+      (build.env("TEST_SCOPE") == "merge" &&
+       build.env("MERGE_TEST_PLAN") =~ /,eval,/) ||
+      (build.env("TEST_SCOPE") == "direct" &&
+       (build.env("TEST_TYPE") == "eval" || build.env("TEST_TYPE") == "eval_ci"))
+    command: "/opt/fastvideo-ci-runner/run-ci"
+    timeout_in_minutes: 90
+    env:
+      TEST_TYPE: "eval_ci"
+    retry:
+      automatic:
+        - exit_status: 128
+          limit: 3
+        - exit_status: -1
+          limit: 2
+    agents:
+      queue: "ci-runner"

.buildkite/scripts/lanes/api_server.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the OpenAI-compatible API lane.
+set -euo pipefail
+
+exec pytest ./fastvideo/tests/entrypoints/test_openai_api_integration.py -vs

.buildkite/scripts/lanes/distillation_dmd.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the distillation-DMD lane.
+set -euo pipefail
+
+exec pytest ./fastvideo/tests/training/distill/test_distill_dmd.py -vs

.buildkite/scripts/lanes/dreamverse.sh

@@ -0,0 +1,87 @@
+#!/usr/bin/env bash
+# DreamVerse needs a GPU for import-time device resolution, but it does not
+# build or exercise fastvideo-kernel. A checksummed Node archive is installed
+# in the disposable Slurm container because the shared CI image is
+# Python/CUDA focused.
+set -euo pipefail
+
+node_version=v22.23.2
+case $(uname -m) in
+  aarch64 | arm64)
+    node_arch=arm64
+    node_archive_sha256=013b59cfd2819703a6f4a14ab891fc46fc2a4e3f5bcd92de3fb4929b43e35b30
+    ;;
+  x86_64 | amd64)
+    node_arch=x64
+    node_archive_sha256=b294a556e639d64338823920e5866c21c02741742d2e1529ee1a225c1ec9252a
+    ;;
+  *)
+    echo "Unsupported architecture for DreamVerse Node runtime: $(uname -m)" >&2
+    exit 2
+    ;;
+esac
+node_archive="node-${node_version}-linux-${node_arch}.tar.gz"
+node_runtime_root=$(mktemp -d -t fastvideo-node.XXXXXX)
+node_archive_path="${node_runtime_root}/${node_archive}"
+node_install_dir="${node_runtime_root}/${node_archive%.tar.gz}"
+curl --proto '=https' --tlsv1.2 --retry 5 --retry-all-errors \
+  --location --fail --silent --show-error \
+  "https://nodejs.org/dist/${node_version}/${node_archive}" \
+  --output "$node_archive_path"
+printf '%s  %s\n' "$node_archive_sha256" "$node_archive_path" | sha256sum --check --status
+tar -xzf "$node_archive_path" -C "$node_runtime_root"
+export PATH="${node_install_dir}/bin:${PATH}"
+node --version
+npm --version
+
+export PYTHONPATH="$(pwd)/apps/dreamverse${PYTHONPATH:+:$PYTHONPATH}"
+pytest apps/dreamverse/dreamverse/tests -q
+
+cd apps/dreamverse/web
+npm ci
+npm run typecheck
+npm test
+machine_arch=$(uname -m)
+if [[ $machine_arch =~ ^(aarch64|arm64)$ ]]; then
+  npx playwright install --with-deps chromium firefox
+else
+  npx playwright install --with-deps chromium webkit firefox
+fi
+
+master_port=${MASTER_PORT:-7959}
+BACKEND_PORT=${BACKEND_PORT:-$((master_port + 50))}
+python -m uvicorn dreamverse.mock_server:app --host 127.0.0.1 --port "$BACKEND_PORT" &
+mock_server_pid=$!
+cleanup() {
+  kill "$mock_server_pid" 2>/dev/null || true
+  wait "$mock_server_pid" 2>/dev/null || true
+}
+trap cleanup EXIT INT TERM
+
+for _ in {1..30}; do
+  curl -fsS "http://127.0.0.1:$BACKEND_PORT/healthz" && break
+  sleep 1
+done
+curl -fsS "http://127.0.0.1:$BACKEND_PORT/healthz"
+
+if [[ $machine_arch =~ ^(aarch64|arm64)$ ]]; then
+  # Playwright WebKit traps before opening a page on Linux ARM64, and its
+  # bundled Chromium lacks the H.264/AAC codecs used by the fMP4 assertions.
+  # Firefox covers every flow, including streaming. Chromium and its mobile
+  # profile still cover all codec-independent UI behavior on GB200.
+  BACKEND_HOST=127.0.0.1 BACKEND_PORT="$BACKEND_PORT" CI=1 \
+    npm run e2e -- --project=firefox
+  BACKEND_HOST=127.0.0.1 BACKEND_PORT="$BACKEND_PORT" CI=1 \
+    npm run e2e -- \
+      --project=chromium \
+      --project=mobile-chromium \
+      --grep-invert='streams, plays, and surfaces a downloadable clip|starts a new project and switches back to the prior session|saved projects persist across a page reload'
+else
+  BACKEND_HOST=127.0.0.1 BACKEND_PORT="$BACKEND_PORT" CI=1 \
+    npm run e2e -- \
+      --project=chromium \
+      --project=webkit \
+      --project=firefox \
+      --project=mobile-safari \
+      --project=mobile-chromium
+fi

.buildkite/scripts/lanes/encoder.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the encoder lane.
+set -euo pipefail
+
+exec pytest ./fastvideo/tests/encoders -vs

.buildkite/scripts/lanes/eval.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the evaluation lane.
+set -euo pipefail
+
+exec pytest ./fastvideo/tests/eval -vs

.buildkite/scripts/lanes/golden_gate.sh

@@ -0,0 +1,35 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the golden-gate lane. Environment (HF_HOME
+# and authentication) is the runner's responsibility.
+set -euo pipefail
+
+golden_root=./fastvideo/tests/golden_gate
+selected=${FASTVIDEO_GOLDEN_TEST_FILES-}
+if [ -z "$selected" ]; then
+  if [ "${TEST_SCOPE:-}" = merge ]; then
+    echo "Missing FASTVIDEO_GOLDEN_TEST_FILES for merge scope" >&2
+    exit 2
+  fi
+  selected=all
+fi
+if [ "$selected" = all ]; then
+  exec pytest "$golden_root" -xvs
+fi
+
+[[ $selected =~ ^test_[a-z0-9_]+\.py(,test_[a-z0-9_]+\.py)*$ ]] || {
+  echo "Invalid FASTVIDEO_GOLDEN_TEST_FILES selection" >&2
+  exit 2
+}
+
+IFS=, read -r -a golden_files <<< "$selected"
+golden_paths=()
+for golden_file in "${golden_files[@]}"; do
+  golden_path="$golden_root/$golden_file"
+  [ -f "$golden_path" ] || {
+    echo "Selected golden test does not exist: $golden_file" >&2
+    exit 2
+  }
+  golden_paths+=("$golden_path")
+done
+
+exec pytest "${golden_paths[@]}" -xvs

.buildkite/scripts/lanes/inference_lora.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the LoRA-inference lane.
+set -euo pipefail
+
+exec pytest ./fastvideo/tests/inference/lora/test_lora_inference_similarity.py -vs

.buildkite/scripts/lanes/inference_vmoba.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the VMoBA-inference lane.
+set -euo pipefail
+
+exec python fastvideo/tests/inference/vmoba/test_vmoba_inference.py

.buildkite/scripts/lanes/kernel_tests.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the custom-kernel lane.
+set -euo pipefail
+
+exec pytest fastvideo-kernel/tests/ -vs

.buildkite/scripts/lanes/lora_extraction.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the LoRA-extraction lane.
+set -euo pipefail
+
+exec pytest ./fastvideo/tests/lora_extraction/test_lora_extraction.py -vs

.buildkite/scripts/lanes/performance.sh

@@ -0,0 +1,58 @@
+#!/usr/bin/env bash
+# Canonical Slurm performance lane. Reports are written outside the checkout
+# so the trusted host driver can upload them after untrusted code exits.
+set -uo pipefail
+
+export PERFORMANCE_TRACKING_ROOT=/tmp/perf-tracking
+export PERF_REPORTS_DIR=/workspace/artifacts/performance
+mkdir -p "$PERF_REPORTS_DIR"
+
+if [[ ${BUILDKITE_PULL_REQUEST:-false} =~ ^[1-9][0-9]*$ ]]; then
+  export PERF_RUN_SOURCE=pr
+  export PERF_UPLOAD_POLICY=pass
+elif [ "${BUILDKITE_BRANCH:-}" = main ] \
+  && { [ "${BUILDKITE_SOURCE:-}" = schedule ] || [ "${TEST_SCOPE:-}" = full ]; }; then
+  export PERF_RUN_SOURCE=scheduled_main
+  export PERF_UPLOAD_POLICY=always
+elif [ "${TEST_SCOPE:-}" = direct ]; then
+  export PERF_RUN_SOURCE=unknown
+  export PERF_UPLOAD_POLICY=pass
+else
+  export PERF_RUN_SOURCE=unknown
+  export PERF_UPLOAD_POLICY=never
+fi
+
+# Alternate GPU backends compare against references without publishing records.
+# Their worker has read-only Hub credentials; publication is an operator task.
+if [ "${FASTVIDEO_CI_LOCAL_ONLY:-0}" = 1 ]; then
+  export PERF_UPLOAD_POLICY=never
+fi
+
+nvidia-smi \
+  --query-gpu=index,timestamp,clocks.sm,clocks.max.sm,power.draw,power.limit,temperature.gpu \
+  --format=csv -l 10 > "$PERF_REPORTS_DIR/gpu_telemetry.csv" 2>/dev/null &
+telemetry_pid=$!
+cleanup() {
+  kill "$telemetry_pid" 2>/dev/null || true
+  wait "$telemetry_pid" 2>/dev/null || true
+}
+trap cleanup EXIT INT TERM
+
+pytest ./fastvideo/tests/performance -vs
+pytest_rc=$?
+compare_rc=0
+if [ "$pytest_rc" -eq 0 ] || [ "$PERF_UPLOAD_POLICY" = always ]; then
+  PERF_PYTEST_RC=$pytest_rc python ./fastvideo/tests/performance/compare_baseline.py
+  compare_rc=$?
+fi
+python ./fastvideo/tests/performance/dashboard.py || true
+cp -f fastvideo/tests/performance/results/*.json "$PERF_REPORTS_DIR/" 2>/dev/null || true
+
+echo "--- GPU telemetry (clocks.sm vs clocks.max.sm reveals capped hosts) ---"
+cat "$PERF_REPORTS_DIR/gpu_telemetry.csv" || true
+
+final_rc=$pytest_rc
+if [ "$final_rc" -eq 0 ]; then
+  final_rc=$compare_rc
+fi
+exit "$final_rc"

.buildkite/scripts/lanes/self_forcing.sh

@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the self-forcing lane.
+set -euo pipefail
+
+export WANDB_MODE=offline
+exec pytest ./fastvideo/tests/training/self-forcing/test_self_forcing.py -vs

.buildkite/scripts/lanes/ssim.sh

@@ -0,0 +1,40 @@
+#!/usr/bin/env bash
+# Canonical four-GPU SSIM lane for the Slinky Slurm worker.
+set -euo pipefail
+
+args=()
+if [ "${FASTVIDEO_SSIM_BOOTSTRAP_MODE:-0}" = 1 ]; then
+  args+=(--bootstrap-mode)
+fi
+selected=${FASTVIDEO_SSIM_TEST_FILES-}
+if [ -z "$selected" ]; then
+  if [ "${TEST_SCOPE:-}" = merge ]; then
+    echo "Missing FASTVIDEO_SSIM_TEST_FILES for merge scope" >&2
+    exit 2
+  fi
+  selected=all
+fi
+if [ "$selected" != all ]; then
+  [[ $selected =~ ^test_[a-z0-9_]+\.py(,test_[a-z0-9_]+\.py)*$ ]] || {
+    echo "Invalid FASTVIDEO_SSIM_TEST_FILES selection" >&2
+    exit 2
+  }
+  IFS=, read -r -a ssim_files <<< "$selected"
+  for ssim_file in "${ssim_files[@]}"; do
+    args+=(--test-file "$ssim_file")
+  done
+fi
+
+# MoGe's utils3d dependency builds glcontext from source on ARM64. The current
+# runner image predates the baked-in X11 headers below, so keep this guarded
+# bootstrap until every deployed image digest contains libx11-dev.
+if [ ! -f /usr/include/X11/Xlib.h ]; then
+  apt-get -o Acquire::Retries=5 update
+  apt-get -o Acquire::Retries=5 install -y --no-install-recommends libx11-dev
+  rm -rf /var/lib/apt/lists/*
+fi
+
+uv pip install git+https://github.com/microsoft/MoGe.git
+uv pip install k_diffusion einops_exts alias_free_torch torchsde
+
+exec python fastvideo/tests/ssim/ci_runner.py "${args[@]}"

.buildkite/scripts/lanes/train_framework.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the modular training-framework lane.
+set -euo pipefail
+
+exec pytest ./fastvideo/tests/train/models ./fastvideo/tests/train/methods -vs

.buildkite/scripts/lanes/training.sh

@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the legacy vanilla-training lane.
+set -euo pipefail
+
+export WANDB_MODE=offline
+exec pytest ./fastvideo/tests/training/Vanilla -srP

.buildkite/scripts/lanes/training_lora.sh

@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the legacy LoRA-training lane.
+set -euo pipefail
+
+export WANDB_MODE=offline
+exec pytest ./fastvideo/tests/training/lora/test_lora_training.py -srP

.buildkite/scripts/lanes/training_vsa.sh

@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the legacy VSA-training lane.
+set -euo pipefail
+
+export WANDB_MODE=offline
+exec pytest ./fastvideo/tests/training/VSA -srP

.buildkite/scripts/lanes/transformer.sh

@@ -0,0 +1,9 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the transformer lane.
+set -euo pipefail
+
+# The existing block reference records an absent FASTVIDEO_FA4 (FA2). Keep
+# that reference identity; the component lane also selects FA2 explicitly.
+env -u FASTVIDEO_FA4 pytest ./fastvideo/tests/golden_gate/test_wan_t2v.py -xvs
+pytest ./fastvideo/tests/golden_gate/test_wan_causal.py -xvs
+exec pytest ./fastvideo/tests/transformers -vs

.buildkite/scripts/lanes/vae.sh

@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Canonical Slurm CI selection for the VAE lane.
+set -euo pipefail
+
+pytest ./fastvideo/tests/golden_gate/test_wan_vae.py -xvs
+exec pytest ./fastvideo/tests/vaes -vs

.buildkite/scripts/pr_test.sh

@@ -1,6 +1,19 @@
 #!/bin/bash
 set -uo pipefail
 
+# DORMANT ROLLBACK ONLY. Active CI is Slurm-only and pipeline.yml never calls
+# this launcher. Refuse every Buildkite invocation even if a stale step or
+# operator typo reaches this file; local rollback experiments require an
+# explicit opt-in.
+if [ -n "${BUILDKITE:-}" ]; then
+  echo "Legacy Modal CI is disabled; use the Slinky Slurm runner." >&2
+  exit 2
+fi
+if [ "${FASTVIDEO_ENABLE_LEGACY_MODAL_CI:-0}" != 1 ]; then
+  echo "Legacy Modal CI is dormant. Set FASTVIDEO_ENABLE_LEGACY_MODAL_CI=1 only for a manual rollback test." >&2
+  exit 2
+fi
+
 log() {
   echo "[$(date '+%Y-%m-%d %H:%M:%S')] $1"
 }
@@ -76,7 +89,7 @@ EFFECTIVE_PR=${BUILDKITE_PULL_REQUEST:-false}
 if [ "$EFFECTIVE_PR" = "false" ] && [ -n "${PR_NUMBER:-}" ]; then
     EFFECTIVE_PR=$PR_NUMBER
 fi
-MODAL_ENV="BUILDKITE_REPO=$BUILDKITE_REPO BUILDKITE_COMMIT=$BUILDKITE_COMMIT BUILDKITE_PULL_REQUEST=$EFFECTIVE_PR BUILDKITE_BRANCH=${BUILDKITE_BRANCH:-} TEST_SCOPE=${TEST_SCOPE:-} BUILDKITE_BUILD_URL=${BUILDKITE_BUILD_URL:-} BUILDKITE_BUILD_ID=${BUILDKITE_BUILD_ID:-} BUILDKITE_JOB_ID=${BUILDKITE_JOB_ID:-} IMAGE_VERSION=$IMAGE_VERSION"
+MODAL_ENV="BUILDKITE_REPO=$BUILDKITE_REPO BUILDKITE_COMMIT=$BUILDKITE_COMMIT BUILDKITE_PULL_REQUEST=$EFFECTIVE_PR BUILDKITE_BRANCH=${BUILDKITE_BRANCH:-} BUILDKITE_SOURCE=${BUILDKITE_SOURCE:-} TEST_SCOPE=${TEST_SCOPE:-} BUILDKITE_BUILD_URL=${BUILDKITE_BUILD_URL:-} BUILDKITE_BUILD_ID=${BUILDKITE_BUILD_ID:-} BUILDKITE_JOB_ID=${BUILDKITE_JOB_ID:-} IMAGE_VERSION=$IMAGE_VERSION"
 
 POST_RUN_HOOK=""
 
@@ -187,6 +200,10 @@ case "$TEST_TYPE" in
         log "Running transformer tests..."
         MODAL_COMMAND="$MODAL_ENV HF_API_KEY=$HF_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_transformer_tests"
         ;;
+    "golden_gate")
+        log "Running golden-gate tests..."
+        MODAL_COMMAND="$MODAL_ENV HF_API_KEY=$HF_API_KEY python3 -m modal run $MODAL_TEST_FILE::run_golden_gate_tests"
+        ;;
     "ssim")
         log "Running SSIM tests..."
         SSIM_BOOTSTRAP_ARGS=$(ssim_bootstrap_args)

.buildkite/scripts/unit_test.sh

@@ -0,0 +1,26 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+exec pytest \
+  ./fastvideo/tests/api/ \
+  ./fastvideo/tests/contract/ \
+  ./fastvideo/tests/dataset/ \
+  ./fastvideo/tests/workflow/ \
+  ./fastvideo/tests/entrypoints/ \
+  ./fastvideo/tests/loader/ \
+  ./fastvideo/tests/pipelines/ \
+  ./fastvideo/tests/platforms/ \
+  ./fastvideo/tests/train/ \
+  ./fastvideo/tests/stages/ \
+  ./fastvideo/tests/ops/ \
+  ./fastvideo/tests/worker/ \
+  ./fastvideo/tests/training/test_trackers.py \
+  ./fastvideo/tests/attention/test_sdpa_metadata_mask_contract.py \
+  ./fastvideo/tests/attention/test_vsa_h3_tile_grad_safety.py \
+  ./fastvideo/tests/modal/test_kernel_build_cache.py \
+  ./fastvideo/tests/modal/test_pr_test.py \
+  ./fastvideo/tests/modal/test_ssim_test.py \
+  --ignore=./fastvideo/tests/entrypoints/test_openai_api_integration.py \
+  --ignore=./fastvideo/tests/train/models \
+  --ignore=./fastvideo/tests/train/methods \
+  -vs

.github/PULL_REQUEST_TEMPLATE.md

@@ -8,10 +8,10 @@ PR TITLE: Must start with a type tag, e.g.:
 MERGE WORKFLOW:
   1. Ensure pre-commit passes and you have at least 1 approval
   2. Comment /merge (or add the "ready" label) to enter the Merge Queue
-  3. Full Test Suite runs automatically on a staging branch → auto-merge on success
+  3. A path-aware merge gate runs only relevant integration tests → auto-merge on success
 
 ON-DEMAND TESTING (write access required):
-  /test full       — Full Test Suite        /test ssim        — SSIM regression
+  /test full       — Explicit all-lane run  /test ssim        — Full SSIM regression
   /test training   — Training pipeline      /test encoder     — Encoder tests
   /test transformer — Transformer tests     /test vae         — VAE tests
   /test kernel     — CUDA kernel tests      /test unit        — Unit tests

.github/scripts/gate_full_suite.sh

@@ -1,14 +1,14 @@
 #!/usr/bin/env bash
-# Gate the expensive Buildkite full suite on the cheap GitHub checks.
+# Gate the path-aware Buildkite merge plan on the cheap GitHub checks.
 #
 # Polls the workflow runs for the PR head commit and only exits 0 once the
 # watched cheap workflows (pre-commit, docs build) have succeeded, so the
-# 'ready' label cannot burn ~20 GPU lanes on a head that a cheap check has
-# already doomed.
+# 'ready' label cannot burn path-selected GPU lanes on a head that a cheap
+# check has already doomed.
 #
 # Semantics:
 #   - watched run completed with a bad conclusion  -> exit 1 (fail CLOSED:
-#     no full suite; the next push re-arms via the 'synchronize' trigger)
+#     no merge gate; the next push re-arms via the 'synchronize' trigger)
 #   - watched run cancelled                         -> still pending: the docs
 #     workflow's repo-global 'pages' concurrency group cancels runs superseded
 #     by unrelated pushes, so 'cancelled' is not a verdict on this PR
@@ -29,7 +29,7 @@ set -euo pipefail
 : "${PR_NUMBER:?PR_NUMBER (pull request number) is required}"
 : "${GITHUB_REPOSITORY:?GITHUB_REPOSITORY is required}"
 
-# Workflow-level `name:` values that must be green before the full suite
+# Workflow-level `name:` values that must be green before the merge gate
 # may start. "Deploy Documentation" is path-filtered on PRs, so its run may
 # legitimately never exist; pre-commit always runs, so it must appear.
 WATCHED_NAMES='["pre-commit", "Deploy Documentation"]'
@@ -56,7 +56,7 @@ recheck_ready_label() {
   if pr_json=$(gh_api "repos/${GITHUB_REPOSITORY}/pulls/${PR_NUMBER}" 2>/dev/null); then
     if ! jq -e '[.labels[]?.name] | index("ready")' <<<"$pr_json" >/dev/null 2>&1; then
       echo "::error::PR #${PR_NUMBER} no longer has the 'ready' label —" \
-        "NOT triggering the Buildkite full suite. Re-add the label to re-arm."
+        "NOT triggering the Buildkite merge gate. Re-add the label to re-arm."
       exit 1
     fi
   else
@@ -84,7 +84,7 @@ while true; do
         | map(.name) | join(", ")' <<<"$state")
     if [ -n "$failed" ]; then
       echo "::error::Cheap check(s) failed on ${PR_SHA}: ${failed}." \
-        "NOT triggering the Buildkite full suite. Push a fix (the 'ready'" \
+        "NOT triggering the Buildkite merge gate. Push a fix (the 'ready'" \
         "label re-arms on every push), or re-run the failed check and then" \
         "re-run this workflow."
       exit 1
@@ -97,7 +97,7 @@ while true; do
     if [ "$pending" -eq 0 ]; then
       if [ -z "$missing" ]; then
         recheck_ready_label
-        echo "All watched cheap checks are green — full suite may proceed."
+        echo "All watched cheap checks are green — merge gate may proceed."
         exit 0
       fi
       case "$missing" in
@@ -119,14 +119,14 @@ while true; do
     echo "::warning::GitHub API error querying workflow runs for ${PR_SHA} (attempt ${api_fails}/3)."
     if [ "$api_fails" -ge 3 ]; then
       recheck_ready_label
-      echo "::warning::FAILING OPEN: cannot query GitHub check status — triggering the full suite WITHOUT the cheap-check gate."
+      echo "::warning::FAILING OPEN: cannot query GitHub check status — triggering the merge gate WITHOUT the cheap-check gate."
       exit 0
     fi
   fi
 
   if [ "$elapsed" -ge "$MAX_WAIT_SECS" ]; then
     recheck_ready_label
-    echo "::warning::FAILING OPEN: watched checks still pending after $(( MAX_WAIT_SECS / 60 )) min${missing:+ (never appeared: ${missing})} — triggering the full suite anyway."
+    echo "::warning::FAILING OPEN: watched checks still pending after $(( MAX_WAIT_SECS / 60 )) min${missing:+ (never appeared: ${missing})} — triggering the merge gate anyway."
     exit 0
   fi
   sleep "$POLL_SECS"

.github/scripts/plan_merge_ci.py

@@ -0,0 +1,582 @@
+#!/usr/bin/env python3
+"""Select the additive GPU integration lanes needed by a PR diff.
+
+Fastcheck is the universal six-lane baseline and is intentionally not repeated
+here.  This planner selects only the more expensive merge-gate lanes.  Unknown
+source/build paths fail closed to the complete integration set, while explicit
+documentation and repository-metadata paths require no additional GPU work.
+"""
+
+from __future__ import annotations
+
+import argparse
+import fnmatch
+import re
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import TextIO
+
+MERGE_LANES = (
+    "golden-gate",
+    "ssim",
+    "lora-inference",
+    "lora-extraction",
+    "training",
+    "distillation",
+    "self-forcing",
+    "lora-training",
+    "training-vsa",
+    "inference-vmoba",
+    "performance",
+    "api-server",
+    "train-framework",
+    "eval",
+)
+
+LANE_SCRIPT_TO_KEY = {
+    "api_server.sh": "api-server",
+    "distillation_dmd.sh": "distillation",
+    "eval.sh": "eval",
+    "golden_gate.sh": "golden-gate",
+    "inference_lora.sh": "lora-inference",
+    "inference_vmoba.sh": "inference-vmoba",
+    "lora_extraction.sh": "lora-extraction",
+    "performance.sh": "performance",
+    "self_forcing.sh": "self-forcing",
+    "ssim.sh": "ssim",
+    "train_framework.sh": "train-framework",
+    "training.sh": "training",
+    "training_lora.sh": "lora-training",
+    "training_vsa.sh": "training-vsa",
+}
+
+FASTCHECK_LANE_SCRIPTS = {
+    "dreamverse.sh",
+    "encoder.sh",
+    "kernel_tests.sh",
+    "transformer.sh",
+    "vae.sh",
+}
+
+LEGACY_TRAINING_LANES = (
+    "training",
+    "distillation",
+    "self-forcing",
+    "lora-training",
+    "training-vsa",
+)
+
+ALL_TRAINING_LANES = (*LEGACY_TRAINING_LANES, "train-framework")
+
+SSIM_SMOKE_TESTS = (
+    "test_flux_t2i_similarity.py",
+    "test_wan_t2v_similarity.py",
+)
+
+SAFE_PATTERNS = (
+    "*.md",
+    "*.rst",
+    ".agents/**",
+    ".claude/**",
+    ".codex/**",
+    ".github/ISSUE_TEMPLATE/**",
+    ".github/PULL_REQUEST_TEMPLATE.md",
+    ".github/dependabot.yml",
+    ".github/mergify.yml",
+    ".github/scripts/**",
+    ".github/workflows/**",
+    ".buildkite/scripts/pre_commit.sh",
+    ".git-blame-ignore-revs",
+    ".gitattributes",
+    ".gitignore",
+    ".pre-commit-config.yaml",
+    "AGENTS.md",
+    "CITATION.cff",
+    "CODE_OF_CONDUCT.md",
+    "CONTRIBUTING.md",
+    "LICENSE",
+    "NOTICE",
+    "__init__.py",
+    "collect_env.py",
+    "SECURITY.md",
+    "assets/**",
+    "comfyui/**",
+    "docs/**",
+    "examples/**",
+    "mkdocs.yml",
+    "requirements-mkdocs.in",
+    "requirements-mkdocs.txt",
+    "scripts/**",
+    "tests/__init__.py",
+    "tests/local_tests/**",
+)
+
+ALL_IMPACT_PATTERNS = (
+    ".buildkite/pipeline.yml",
+    "docker/**",
+    "pyproject.toml",
+    "requirements*.txt",
+    "setup.cfg",
+    "setup.py",
+    "uv.lock",
+)
+
+
+@dataclass(frozen=True)
+class FamilyCoverage:
+    pattern: re.Pattern[str]
+    golden_tests: tuple[str, ...]
+    ssim_tests: tuple[str, ...]
+
+
+FAMILY_COVERAGE = (
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])dreamx(_world)?([/_.-]|$)"),
+        ("test_dreamx.py", ),
+        ("test_dreamx_world_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])flux[_-]?2([/_.-]|$)"),
+        ("test_flux2_klein.py", ),
+        ("test_flux2_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])flux(?![_-]?2)([/_.-]|$)"),
+        ("test_flux.py", ),
+        ("test_flux_t2i_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])(hunyuan)?gamecraft([/_.-]|$)"),
+        ("test_gamecraft.py", ),
+        ("test_gamecraft_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])gen3c([/_.-]|$)"),
+        ("test_gen3c.py", ),
+        ("test_gen3c_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])glm[_-]?image([/_.-]|$)"),
+        ("test_glm_image.py", ),
+        ("test_glm_image_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])kandinsky[_-]?5([/_.-]|$)"),
+        ("test_kandinsky5.py", ),
+        ("test_kandinsky5_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])lingbot([a-z0-9_-]*)([/_.-]|$)"),
+        ("test_lingbot.py", ),
+        ("test_lingbot_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])longcat([/_.-]|$)"),
+        ("test_longcat.py", ),
+        ("test_longcat_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])ltx[_-]?2([/_.-]|$)"),
+        ("test_ltx2.py", ),
+        ("test_ltx2_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])matrixgame[_-]?2([/_.-]|$)"),
+        ("test_matrixgame.py", ),
+        ("test_matrixgame2_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])matrixgame[_-]?3([/_.-]|$)"),
+        ("test_matrixgame.py", ),
+        ("test_matrixgame3_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])minimax[_-]?h3([/_.-]|$)"),
+        ("test_minimax_h3_t2v.py", ),
+        ("test_minimax_h3_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])sd[_-]?3([._-]?5)?([/_.-]|$)"),
+        ("test_sd35.py", ),
+        ("test_sd35_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])stable[_-]?audio([/_.-]|$)"),
+        ("test_stable_audio.py", ),
+        ("test_stable_audio_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])turbo(diffusion)?([/_.-]|$)"),
+        (),
+        ("test_turbodiffusion_similarity.py", ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])wan(video|vae)?([/_.-]|$)"),
+        ("test_wan_t2v.py", "test_wan_vae.py", "test_wan_causal.py", "test_wan_denoising.py"),
+        (
+            "test_causal_similarity.py",
+            "test_wan_i2v_similarity.py",
+            "test_wan_t2v_similarity.py",
+        ),
+    ),
+    FamilyCoverage(
+        re.compile(r"(^|[/_.-])z[_-]?image([/_.-]|$)"),
+        ("test_zimage.py", ),
+        ("test_zimage_similarity.py", ),
+    ),
+)
+
+
+@dataclass
+class MergePlan:
+    lanes: set[str] = field(default_factory=set)
+    golden_tests: set[str] = field(default_factory=set)
+    ssim_tests: set[str] = field(default_factory=set)
+    golden_all: bool = False
+    ssim_all: bool = False
+    reasons: list[str] = field(default_factory=list)
+
+    def add_lanes(self, *lanes: str, reason: str) -> None:
+        unknown = set(lanes) - set(MERGE_LANES)
+        if unknown:
+            raise ValueError(f"Unknown merge lanes: {sorted(unknown)}")
+        self.lanes.update(lanes)
+        self.reasons.append(reason)
+
+    def add_golden(self, tests: tuple[str, ...], reason: str) -> None:
+        self.add_lanes("golden-gate", reason=reason)
+        self.golden_tests.update(tests)
+
+    def add_ssim(self, tests: tuple[str, ...], reason: str) -> None:
+        self.add_lanes("ssim", reason=reason)
+        self.ssim_tests.update(tests)
+
+    def require_all(self, reason: str) -> None:
+        self.lanes.update(MERGE_LANES)
+        self.golden_all = True
+        self.ssim_all = True
+        self.reasons.append(reason)
+
+    def ordered_lanes(self) -> tuple[str, ...]:
+        return tuple(lane for lane in MERGE_LANES if lane in self.lanes)
+
+    def encoded_lanes(self) -> str:
+        lanes = self.ordered_lanes()
+        return "," + ",".join(lanes or ("none", )) + ","
+
+    def encoded_golden_tests(self) -> str:
+        if "golden-gate" not in self.lanes:
+            return "none"
+        if self.golden_all or not self.golden_tests:
+            return "all"
+        return ",".join(sorted(self.golden_tests))
+
+    def encoded_ssim_tests(self) -> str:
+        if "ssim" not in self.lanes:
+            return "none"
+        if self.ssim_all or not self.ssim_tests:
+            return "all"
+        return ",".join(sorted(self.ssim_tests))
+
+
+def _matches_any(path: str, patterns: tuple[str, ...]) -> bool:
+    return any(fnmatch.fnmatchcase(path, pattern) for pattern in patterns)
+
+
+def _family_coverage(path: str) -> tuple[set[str], set[str]]:
+    normalized = path.lower()
+    golden: set[str] = set()
+    ssim: set[str] = set()
+    for family in FAMILY_COVERAGE:
+        if family.pattern.search(normalized):
+            golden.update(family.golden_tests)
+            ssim.update(family.ssim_tests)
+    # Select the component actually touched, including compatibility paths.
+    # Family configs/pipeline wiring can affect all four Wan gates.
+    if re.search(r"(^|[/_.-])wan(video|vae)?([/_.-]|$)", normalized):
+        if (normalized.endswith(("/wan/vae.py", "/wan/vae_config.py", "/vaes/wanvae.py"))
+                or normalized.endswith("/wan/stages/conditioning.py")):
+            golden = {"test_wan_vae.py"}
+        elif normalized.endswith(("/wan/causal_transformer.py", "/dits/causal_wanvideo.py",
+                                  "/wan/stages/causal_denoising.py")):
+            golden = {"test_wan_causal.py"}
+        elif (normalized == "fastvideo/models/dits/wanvideo.py"
+              or normalized.endswith(("/wan/transformer.py", "/wan/stages/denoising.py", "/wan/stages/dmd.py"))):
+            golden = {"test_wan_t2v.py", "test_wan_denoising.py"}
+    return golden, ssim
+
+
+def _select_output_coverage(plan: MergePlan, path: str) -> None:
+    golden, ssim = _family_coverage(path)
+    if golden:
+        plan.add_golden(tuple(sorted(golden)), reason=f"model-family golden coverage: {path}")
+    else:
+        plan.golden_all = True
+        plan.add_lanes("golden-gate", reason=f"shared output golden coverage: {path}")
+    if ssim:
+        plan.add_ssim(tuple(sorted(ssim)), reason=f"model-family SSIM coverage: {path}")
+    else:
+        plan.add_ssim(SSIM_SMOKE_TESTS, reason=f"shared output SSIM smoke coverage: {path}")
+
+
+def classify_paths(paths: list[str]) -> MergePlan:
+    plan = MergePlan()
+    normalized_paths: list[str] = []
+    for raw_path in paths:
+        path = raw_path.strip()
+        while path.startswith("./"):
+            path = path[2:]
+        if path:
+            normalized_paths.append(path)
+    normalized_paths = sorted(set(normalized_paths))
+    if not normalized_paths:
+        plan.require_all("changed-file list was empty; failing closed")
+        return plan
+
+    for path in normalized_paths:
+        if path == "__FASTVIDEO_CI_PLAN_ALL__":
+            plan.require_all("changed-file API failed; failing closed")
+            continue
+
+        if path in {"requirements-mkdocs.in", "requirements-mkdocs.txt"}:
+            plan.reasons.append(f"documentation dependencies need no GPU integration: {path}")
+            continue
+
+        if _matches_any(path, ALL_IMPACT_PATTERNS):
+            plan.require_all(f"cross-cutting build/runtime surface: {path}")
+            continue
+
+        lane_script_prefix = ".buildkite/scripts/lanes/"
+        if path.startswith(lane_script_prefix):
+            script_name = Path(path).name
+            lane = LANE_SCRIPT_TO_KEY.get(script_name)
+            if lane is None:
+                if script_name in FASTCHECK_LANE_SCRIPTS:
+                    plan.reasons.append(f"covered by automatic Fastcheck lane: {path}")
+                else:
+                    plan.require_all(f"unknown lane script: {path}")
+            elif lane == "golden-gate":
+                plan.golden_all = True
+                plan.add_lanes(lane, reason=f"golden lane implementation: {path}")
+            elif lane == "ssim":
+                plan.ssim_all = True
+                plan.add_lanes(lane, reason=f"SSIM lane implementation: {path}")
+            else:
+                plan.add_lanes(lane, reason=f"lane implementation: {path}")
+            continue
+
+        if path.startswith("fastvideo/tests/golden_gate/"):
+            name = Path(path).name
+            if name.startswith("test_") and name.endswith(".py"):
+                plan.add_golden((name, ), reason=f"changed golden test: {path}")
+            elif name in {"AGENTS.md", "README.md"}:
+                plan.reasons.append(f"golden documentation only: {path}")
+            else:
+                plan.golden_all = True
+                plan.add_lanes("golden-gate", reason=f"shared golden harness/reference: {path}")
+            continue
+
+        if path.startswith("fastvideo/tests/ssim/"):
+            name = Path(path).name
+            if name.startswith("test_") and name.endswith(".py"):
+                plan.add_ssim((name, ), reason=f"changed SSIM test: {path}")
+            elif path.endswith((".py", ".json", ".pt", ".png", ".mp4")):
+                plan.ssim_all = True
+                plan.add_lanes("ssim", reason=f"shared SSIM harness/reference: {path}")
+            continue
+
+        if path.startswith("fastvideo/tests/performance/") or path.startswith(".buildkite/performance-benchmarks/"):
+            plan.add_lanes("performance", reason=f"performance coverage: {path}")
+            continue
+        if path.startswith(("fastvideo/performance/", "fastvideo/performance_dashboard/",
+                            "apps/performance_dashboard/")):
+            plan.add_lanes("performance", reason=f"performance implementation: {path}")
+            continue
+        if path.startswith("fastvideo/benchmarks/"):
+            if "/mlx_" in path or Path(path).name.startswith("mlx_"):
+                plan.reasons.append(f"covered by the path-filtered macOS MLX workflow: {path}")
+            else:
+                plan.add_lanes("performance", reason=f"benchmark implementation: {path}")
+            continue
+        if path.startswith("fastvideo/tests/eval/") or path.startswith("fastvideo/eval/"):
+            plan.add_lanes("eval", reason=f"evaluation coverage: {path}")
+            continue
+        if path.startswith("fastvideo/third_party/eval/"):
+            plan.add_lanes("eval", reason=f"vendored evaluation implementation: {path}")
+            continue
+        if path.startswith("fastvideo/tests/lora_extraction/") or path.startswith("scripts/lora_extraction/"):
+            plan.add_lanes("lora-extraction", reason=f"LoRA extraction coverage: {path}")
+            continue
+        if path.startswith("fastvideo/tests/inference/lora/"):
+            plan.add_lanes("lora-inference", reason=f"LoRA inference coverage: {path}")
+            continue
+        if path.startswith("fastvideo/tests/inference/vmoba/"):
+            plan.add_lanes("inference-vmoba", reason=f"VMoBA inference coverage: {path}")
+            continue
+        if path.startswith(("fastvideo/dataset/", "fastvideo/workflow/", "fastvideo/pipelines/preprocess/",
+                            "fastvideo/pipelines/training/")):
+            plan.add_lanes(*ALL_TRAINING_LANES, reason=f"shared data/training input surface: {path}")
+            continue
+        if path.startswith("fastvideo/tests/train/") or path.startswith("fastvideo/train/"):
+            plan.add_lanes("train-framework", reason=f"modular training coverage: {path}")
+            continue
+
+        if path.startswith("fastvideo/tests/training/"):
+            lowered = path.lower()
+            if "/vanilla/" in lowered:
+                plan.add_lanes("training", reason=f"vanilla training coverage: {path}")
+            elif "/distill/" in lowered:
+                plan.add_lanes("distillation", reason=f"distillation coverage: {path}")
+            elif "/self-forcing/" in lowered:
+                plan.add_lanes("self-forcing", reason=f"self-forcing coverage: {path}")
+            elif "/lora/" in lowered:
+                plan.add_lanes("lora-training", reason=f"LoRA training coverage: {path}")
+            elif "/vsa/" in lowered:
+                plan.add_lanes("training-vsa", reason=f"VSA training coverage: {path}")
+            else:
+                plan.add_lanes(*LEGACY_TRAINING_LANES, reason=f"shared legacy training coverage: {path}")
+            continue
+
+        if path.startswith("fastvideo/training/"):
+            lowered = path.lower()
+            if "self_forcing" in lowered:
+                plan.add_lanes("self-forcing", reason=f"self-forcing implementation: {path}")
+            elif "distill" in lowered:
+                plan.add_lanes("distillation", reason=f"distillation implementation: {path}")
+            elif "lora" in lowered:
+                plan.add_lanes("lora-training", reason=f"LoRA training implementation: {path}")
+            else:
+                plan.add_lanes(*LEGACY_TRAINING_LANES, reason=f"shared legacy training implementation: {path}")
+            continue
+
+        lowered = path.lower()
+        if "vmoba" in lowered and path.startswith(("fastvideo/", ".buildkite/")):
+            plan.add_lanes("inference-vmoba", reason=f"VMoBA implementation: {path}")
+            plan.add_golden(("test_wan_t2v.py", ), reason=f"VMoBA end-to-end coverage: {path}")
+            continue
+        if "lora" in lowered and path.startswith("fastvideo/"):
+            plan.add_lanes(
+                "lora-inference",
+                "lora-extraction",
+                "lora-training",
+                reason=f"shared LoRA implementation: {path}",
+            )
+            _select_output_coverage(plan, path)
+            continue
+
+        if path.startswith("fastvideo/entrypoints/") or path.startswith("fastvideo/api/"):
+            plan.add_lanes("api-server", reason=f"API/entrypoint integration: {path}")
+            if "openai" not in lowered and "/cli/" not in lowered:
+                _select_output_coverage(plan, path)
+            continue
+        if path.startswith("fastvideo/worker/"):
+            plan.add_lanes("api-server", reason=f"worker/API integration: {path}")
+            _select_output_coverage(plan, path)
+            continue
+        if path.startswith("fastvideo/distributed/"):
+            plan.add_lanes(
+                "training",
+                "train-framework",
+                reason=f"distributed runtime integration: {path}",
+            )
+            _select_output_coverage(plan, path)
+            continue
+        if path.startswith(("fastvideo/hooks/", "fastvideo/platforms/", "fastvideo/third_party/")):
+            _select_output_coverage(plan, path)
+            continue
+        if path.startswith(("fastvideo/models/", "fastvideo/pipelines/", "fastvideo/configs/",
+                            "fastvideo/layers/", "fastvideo/attention/")):
+            _select_output_coverage(plan, path)
+            continue
+        if path in {
+                "fastvideo/fastvideo_args.py",
+                "fastvideo/forward_context.py",
+                "fastvideo/image_processor.py",
+                "fastvideo/registry.py",
+                "fastvideo/utils.py",
+        }:
+            _select_output_coverage(plan, path)
+            continue
+        if path.startswith("fastvideo/mlx_runtime/"):
+            plan.reasons.append(f"covered by the path-filtered macOS MLX workflow: {path}")
+            continue
+        if path.startswith("fastvideo/logging_utils/") or path in {
+                "fastvideo/__init__.py",
+                "fastvideo/envs.py",
+                "fastvideo/logger.py",
+                "fastvideo/profiler.py",
+                "fastvideo/version.py",
+        }:
+            plan.reasons.append(f"covered by automatic Fastcheck: {path}")
+            continue
+        if path.startswith(("fastvideo-kernel/", "csrc/")):
+            plan.add_golden(("test_wan_t2v.py", ), reason=f"kernel integration smoke: {path}")
+            plan.add_ssim(("test_wan_t2v_similarity.py", ), reason=f"kernel numerical smoke: {path}")
+            continue
+
+        if path.startswith("apps/dreamverse/"):
+            # DreamVerse is already one of the six automatic Fastcheck lanes.
+            plan.reasons.append(f"covered by automatic DreamVerse Fastcheck: {path}")
+            continue
+        if path.startswith("fastvideo/tests/"):
+            # The automatic unit/component Fastcheck lanes own the remaining
+            # package tests. Domain-specific expensive test roots were handled
+            # above.
+            plan.reasons.append(f"covered by automatic Fastcheck: {path}")
+            continue
+        if path in {".buildkite/scripts/unit_test.sh", ".buildkite/scripts/pr_test.sh"}:
+            plan.reasons.append(f"covered by automatic unit Fastcheck: {path}")
+            continue
+        if _matches_any(path, SAFE_PATTERNS):
+            plan.reasons.append(f"no additional GPU integration needed: {path}")
+            continue
+
+        plan.require_all(f"unclassified path; failing closed: {path}")
+
+    return plan
+
+
+def _write_github_output(output: TextIO, plan: MergePlan) -> None:
+    output.write(f"merge_test_plan={plan.encoded_lanes()}\n")
+    output.write(f"merge_golden_tests={plan.encoded_golden_tests()}\n")
+    output.write(f"merge_ssim_tests={plan.encoded_ssim_tests()}\n")
+    output.write(f"merge_plan_label={','.join(plan.ordered_lanes()) or 'none'}\n")
+
+
+def _write_summary(output: TextIO, plan: MergePlan) -> None:
+    output.write("## Change-aware merge test plan\n\n")
+    output.write("| Selection | Value |\n|---|---|\n")
+    output.write(f"| Additional Slurm lanes | `{','.join(plan.ordered_lanes()) or 'none'}` |\n")
+    output.write(f"| Golden tests | `{plan.encoded_golden_tests()}` |\n")
+    output.write(f"| SSIM tests | `{plan.encoded_ssim_tests()}` |\n\n")
+    output.write("Fastcheck remains the universal six-lane baseline.\n")
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--paths-file", type=Path, required=True)
+    parser.add_argument("--github-output", type=Path)
+    parser.add_argument("--summary-file", type=Path)
+    return parser.parse_args()
+
+
+def main() -> int:
+    args = parse_args()
+    paths = args.paths_file.read_text(encoding="utf-8").splitlines()
+    plan = classify_paths(paths)
+    print(f"MERGE_TEST_PLAN={plan.encoded_lanes()}")
+    print(f"MERGE_GOLDEN_TESTS={plan.encoded_golden_tests()}")
+    print(f"MERGE_SSIM_TESTS={plan.encoded_ssim_tests()}")
+    for reason in plan.reasons:
+        print(f"- {reason}")
+    if args.github_output:
+        with args.github_output.open("a", encoding="utf-8") as output:
+            _write_github_output(output, plan)
+    if args.summary_file:
+        with args.summary_file.open("a", encoding="utf-8") as output:
+            _write_summary(output, plan)
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

.github/scripts/test_gate_full_suite.sh

@@ -53,7 +53,7 @@ PC_PENDING='{"name": "pre-commit", "id": 1, "status": "in_progress", "conclusion
 DOCS_OK='{"name": "Deploy Documentation", "id": 2, "status": "completed", "conclusion": "success"}'
 DOCS_BAD='{"name": "Deploy Documentation", "id": 2, "status": "completed", "conclusion": "failure"}'
 DOCS_CANCELLED='{"name": "Deploy Documentation", "id": 2, "status": "completed", "conclusion": "cancelled"}'
-OTHER='{"name": "Trigger Full Suite", "id": 3, "status": "in_progress", "conclusion": null}'
+OTHER='{"name": "Trigger Merge Gate", "id": 3, "status": "in_progress", "conclusion": null}'
 NULL_NAME='{"name": null, "id": 4, "status": "completed", "conclusion": "failure"}'
 PC_OK_RERUN='{"name": "pre-commit", "id": 5, "status": "completed", "conclusion": "success"}'
 

.github/workflows/_template-build-image.yml

@@ -190,6 +190,7 @@ jobs:
         if: ${{ !inputs.push_by_digest }}
         run: |
           echo "✅ Python ${{ inputs.python_version }} image successfully built and pushed to ${{ steps.image.outputs.name }}:${{ inputs.tag_suffix }}-sha-${GITHUB_SHA::7}"
+          echo "Digest: ${{ steps.build-push.outputs.digest }}"
           echo "To run tests with this image, manually trigger the 'Run Tests' workflow."
 
       - name: Digest success message

.github/workflows/ci-aggregate-status.yml

@@ -11,6 +11,7 @@ jobs:
     if: >-
       github.event.context == 'direct-test-completed'
       && github.event.state == 'success'
+      && (vars.CI_GPU_BACKEND == '' || vars.CI_GPU_BACKEND == 'slurm')
     runs-on: ubuntu-latest
     steps:
       - name: Check and update aggregate status
@@ -26,29 +27,48 @@ jobs:
               per_page: 100,
             });
 
-            const bkStatuses = data.statuses.filter(
-              s => s.context.startsWith('buildkite/ci/')
-            );
-
-            const FASTCHECK_PREFIX = 'buildkite/ci/microscope-';
+            // Buildkite derives the GitHub context prefix from the label emoji.
+            // Keep hard Full Suite lanes in test-tube/bar-chart namespaces and
+            // Fastcheck lanes in microscope so targeted reruns cannot clear the
+            // wrong aggregate status. Automatic PR jobs use pr-fastcheck while
+            // slash-command and Full Suite jobs use ci; normalize the suffix
+            // and keep the newest status for each logical lane.
+            const FASTCHECK_PREFIXES = [
+              'buildkite/pr-fastcheck/microscope-',
+              'buildkite/ci/microscope-',
+            ];
             const FULL_SUITE_PREFIXES = [
               'buildkite/ci/test-tube-',
               'buildkite/ci/bar-chart-',
             ];
 
-            const fastcheck = bkStatuses.filter(
-              s => s.context.startsWith(FASTCHECK_PREFIX)
-            );
-            const fullSuite = bkStatuses.filter(
-              s => FULL_SUITE_PREFIXES.some(p => s.context.startsWith(p))
-            );
+            function newestByLane(prefixes) {
+              const statuses = new Map();
+              for (const status of data.statuses) {
+                const prefix = prefixes.find(p => status.context.startsWith(p));
+                if (!prefix) continue;
+                const lane = status.context.slice(prefix.length);
+                const previous = statuses.get(lane);
+                if (!previous || Date.parse(status.updated_at) > Date.parse(previous.updated_at)) {
+                  statuses.set(lane, status);
+                }
+              }
+              return statuses;
+            }
 
-            if (
-              fastcheck.length > 0
-              && fastcheck.every(s => s.state === 'success')
-            ) {
+            const fastcheck = newestByLane(FASTCHECK_PREFIXES);
+            const fullSuiteOnly = newestByLane(FULL_SUITE_PREFIXES);
+            const fastcheckPassed =
+              fastcheck.size === 6
+              && [...fastcheck.values()].every(s => s.state === 'success');
+            const fullSuitePassed =
+              fastcheckPassed
+              && fullSuiteOnly.size === 14
+              && [...fullSuiteOnly.values()].every(s => s.state === 'success');
+
+            if (fastcheckPassed) {
               core.info(
-                `All ${fastcheck.length} fastcheck tests passed — updating fastcheck-passed`
+                `All ${fastcheck.size} fastcheck tests passed — updating fastcheck-passed`
               );
               await github.rest.repos.createCommitStatus({
                 owner: context.repo.owner,
@@ -56,17 +76,13 @@ jobs:
                 sha,
                 state: 'success',
                 context: 'fastcheck-passed',
-                description:
-                  `All ${fastcheck.length} fastcheck tests passed`,
+                description: `All ${fastcheck.size} fastcheck tests passed`,
               });
             }
 
-            if (
-              fullSuite.length > 0
-              && fullSuite.every(s => s.state === 'success')
-            ) {
+            if (fullSuitePassed) {
               core.info(
-                `All ${fullSuite.length} full suite tests passed — updating full-suite-passed`
+                'All 20 full suite tests passed — updating full-suite-passed'
               );
               await github.rest.repos.createCommitStatus({
                 owner: context.repo.owner,
@@ -74,7 +90,6 @@ jobs:
                 sha,
                 state: 'success',
                 context: 'full-suite-passed',
-                description:
-                  `All ${fullSuite.length} full suite tests passed`,
+                description: 'All 20 full suite tests passed',
               });
             }

.github/workflows/ci-gpu-backend-status.yml

@@ -0,0 +1,62 @@
+name: Promote Selected GPU Backend Status
+
+on:
+  status:
+
+permissions:
+  statuses: write
+
+concurrency:
+  group: gpu-ci-status-${{ github.event.sha }}-${{ vars.CI_GPU_BACKEND }}
+  cancel-in-progress: false
+
+jobs:
+  promote:
+    if: >-
+      (vars.CI_GPU_BACKEND == 'modal' || vars.CI_GPU_BACKEND == 'vllm')
+      && (github.event.context == format('gpu-ci/{0}/fastcheck-passed', vars.CI_GPU_BACKEND)
+          || github.event.context == format('gpu-ci/{0}/full-suite-passed', vars.CI_GPU_BACKEND))
+    runs-on: ubuntu-latest
+    env:
+      SELECTED_BACKEND: ${{ vars.CI_GPU_BACKEND }}
+    steps:
+      - name: Mirror the selected backend's latest suite results
+        uses: actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea # v7.0.1
+        with:
+          script: |
+            const backend = process.env.SELECTED_BACKEND;
+            if (!['modal', 'vllm'].includes(backend)) {
+              throw new Error('Unsupported selected GPU backend');
+            }
+            const sha = context.payload.sha;
+            // Read current state after entering the serialized workflow. A
+            // delayed event must not overwrite a newer failure with success.
+            const statuses = await github.paginate(github.rest.repos.listCommitStatusesForRef, {
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              ref: sha,
+              per_page: 100,
+            });
+            for (const suffix of ['fastcheck-passed', 'full-suite-passed']) {
+              const sourceContext = `gpu-ci/${backend}/${suffix}`;
+              const matches = statuses.filter(status => status.context === sourceContext);
+              matches.sort((a, b) =>
+                Date.parse(b.updated_at) - Date.parse(a.updated_at) || b.id - a.id
+              );
+              const latest = matches[0];
+              const state = latest ? latest.state : 'pending';
+              if (!['pending', 'success', 'failure', 'error'].includes(state)) {
+                throw new Error(`Unsupported status state for ${sourceContext}`);
+              }
+              await github.rest.repos.createCommitStatus({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                sha,
+                context: suffix,
+                state,
+                description: latest
+                  ? `${backend} ${suffix}: ${state}`
+                  : `Waiting for ${backend} ${suffix}`,
+                ...(latest && latest.target_url ? {target_url: latest.target_url} : {}),
+              });
+            }

.github/workflows/ci-macos-mlx.yml

@@ -0,0 +1,169 @@
+name: macOS MLX Smoke
+
+on:
+  pull_request:
+    branches: [main]
+    paths:
+      - ".github/workflows/ci-macos-mlx.yml"
+      - "fastvideo/mlx_runtime/**"
+      - "fastvideo/tests/mlx/**"
+      - "fastvideo/tests/platforms/test_mps_vsa_error.py"
+      - "fastvideo/tests/platforms/test_cpu_sdpa.py"
+      - "fastvideo/platforms/cpu.py"
+      - "fastvideo/platforms/mps.py"
+      - "fastvideo/platforms/__init__.py"
+      - "fastvideo/__init__.py"
+      - "examples/inference/basic/mlx_*.py"
+      - "fastvideo/benchmarks/mlx_*.py"
+      - "pyproject.toml"
+  workflow_dispatch:
+
+permissions:
+  contents: read
+
+concurrency:
+  group: macos-mlx-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  mlx-smoke:
+    if: github.event_name == 'workflow_dispatch' || github.event.pull_request.draft != true
+    runs-on: macos-15
+    timeout-minutes: 25
+    env:
+      FASTVIDEO_ATTENTION_BACKEND: TORCH_SDPA
+      TOKENIZERS_PARALLELISM: "false"
+      MASTER_ADDR: localhost
+      MASTER_PORT: "29513"
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+          cache: pip
+
+      - uses: astral-sh/setup-uv@v3
+
+      - name: Install lightweight MLX smoke dependencies
+        run: |
+          uv pip install --system \
+            --index-url https://download.pytorch.org/whl/cpu \
+            torch==2.11.0 torchvision torchaudio
+          uv pip install --system \
+            pytest numpy scipy pillow imageio einops cloudpickle filelock \
+            PyYAML diffusers huggingface_hub remote-pdb safetensors loguru mlx \
+            "ftfy>=6.3.1" "opencv-python>=4.10.0.84" psutil "transformers>=5.0.0"
+
+      - name: Show Apple runtime
+        run: |
+          python - <<'PY'
+          import platform
+          import mlx.core as mx
+          import torch
+
+          print("machine:", platform.machine())
+          print("processor:", platform.processor())
+          print("mlx default device:", mx.default_device())
+          memory_size = mx.metal.device_info().get("memory_size") if mx.metal.is_available() else "metal unavailable"
+          print("mlx memory_size:", memory_size)
+          print("torch:", torch.__version__)
+          print("torch mps available:", torch.backends.mps.is_available())
+          PY
+
+      - name: Run MLX smoke tests
+        run: |
+          python -m pytest \
+            fastvideo/tests/mlx/test_dmd_sampling.py \
+            fastvideo/tests/mlx/test_memory_limits.py \
+            fastvideo/tests/mlx/test_quant_capability.py \
+            fastvideo/tests/mlx/test_mlx_dit_parity.py \
+            fastvideo/tests/mlx/test_mlx_compile_parity.py \
+            fastvideo/tests/mlx/test_mlx_checkpoint.py \
+            fastvideo/tests/mlx/test_mlx_checkpoint_compat.py \
+            fastvideo/tests/mlx/test_mlx_affine_dq_gemm.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_parity.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_vsa.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_vsa_regressions.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_fast_mode.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_fast_spatial.py \
+            fastvideo/tests/mlx/test_mlx_fastwan_benchmark.py \
+            fastvideo/tests/mlx/test_taehv_decode.py \
+            fastvideo/tests/mlx/test_frame_upsample.py \
+            fastvideo/tests/mlx/test_mlx_fast_spatial.py \
+            fastvideo/tests/mlx/test_mlx_refine.py \
+            fastvideo/tests/mlx/test_mlx_prompt_enhance.py \
+            fastvideo/tests/mlx/test_mlx_prompt_to_video_decode.py \
+            fastvideo/tests/mlx/test_mlx_wan22_prompt_cache_fingerprint.py \
+            fastvideo/tests/mlx/test_wan22_sample.py \
+            fastvideo/tests/mlx/test_windowed_attention.py \
+            fastvideo/tests/mlx/test_mlx_rife_interpolation.py::test_rife_download_unavailable_has_specific_error \
+            fastvideo/tests/mlx/test_mlx_rife_interpolation.py::test_rife_backend_regression_is_not_skip_eligible \
+            fastvideo/tests/platforms/test_mps_vsa_error.py \
+            fastvideo/tests/platforms/test_cpu_sdpa.py \
+            -v -s -o faulthandler_timeout=120
+
+  # Same tests on MLX's CPU backend. Hosted macOS runners are scarce and
+  # slower to schedule; this Linux job gives fast PR signal on the identical
+  # graph (the parity tests were designed to be backend-agnostic), while the
+  # macOS job above stays the source of truth for Metal behavior.
+  mlx-smoke-linux-cpu:
+    if: github.event_name == 'workflow_dispatch' || github.event.pull_request.draft != true
+    runs-on: ubuntu-latest
+    timeout-minutes: 20
+    env:
+      FASTVIDEO_ATTENTION_BACKEND: TORCH_SDPA
+      TOKENIZERS_PARALLELISM: "false"
+      MASTER_ADDR: localhost
+      MASTER_PORT: "29513"
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+          cache: pip
+
+      - uses: astral-sh/setup-uv@v3
+
+      - name: Install lightweight MLX smoke dependencies (CPU backend)
+        run: |
+          uv pip install --system \
+            --index-url https://download.pytorch.org/whl/cpu \
+            torch==2.11.0 torchvision torchaudio
+          uv pip install --system \
+            pytest numpy scipy pillow imageio einops cloudpickle filelock \
+            PyYAML diffusers huggingface_hub remote-pdb safetensors loguru "mlx[cpu]" \
+            "ftfy>=6.3.1" "opencv-python>=4.10.0.84" psutil "transformers>=5.0.0"
+
+      - name: Run MLX smoke tests (CPU backend)
+        run: |
+          python -m pytest \
+            fastvideo/tests/mlx/test_dmd_sampling.py \
+            fastvideo/tests/mlx/test_memory_limits.py \
+            fastvideo/tests/mlx/test_quant_capability.py \
+            fastvideo/tests/mlx/test_mlx_dit_parity.py \
+            fastvideo/tests/mlx/test_mlx_compile_parity.py \
+            fastvideo/tests/mlx/test_mlx_checkpoint.py \
+            fastvideo/tests/mlx/test_mlx_checkpoint_compat.py \
+            fastvideo/tests/mlx/test_mlx_affine_dq_gemm.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_parity.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_vsa.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_vsa_regressions.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_fast_mode.py \
+            fastvideo/tests/mlx/test_mlx_minimax_h3_fast_spatial.py \
+            fastvideo/tests/mlx/test_mlx_fastwan_benchmark.py \
+            fastvideo/tests/mlx/test_taehv_decode.py \
+            fastvideo/tests/mlx/test_frame_upsample.py \
+            fastvideo/tests/mlx/test_mlx_fast_spatial.py \
+            fastvideo/tests/mlx/test_mlx_refine.py \
+            fastvideo/tests/mlx/test_mlx_prompt_enhance.py \
+            fastvideo/tests/mlx/test_mlx_prompt_to_video_decode.py \
+            fastvideo/tests/mlx/test_mlx_wan22_prompt_cache_fingerprint.py \
+            fastvideo/tests/mlx/test_wan22_sample.py \
+            fastvideo/tests/mlx/test_windowed_attention.py \
+            fastvideo/tests/mlx/test_mlx_rife_interpolation.py::test_rife_download_unavailable_has_specific_error \
+            fastvideo/tests/mlx/test_mlx_rife_interpolation.py::test_rife_backend_regression_is_not_skip_eligible \
+            fastvideo/tests/platforms/test_mps_vsa_error.py \
+            fastvideo/tests/platforms/test_cpu_sdpa.py \
+            -v -s -o faulthandler_timeout=120

.github/workflows/ci-precommit.yml

@@ -27,14 +27,25 @@ jobs:
         ref: ${{ inputs.ref || '' }}
     # For PR events, lint the PR head — but keep the hook definitions from
     # the base branch so an untrusted PR cannot alter what gets executed.
-    - name: Save trusted hook config
+    # The gate scripts are saved too: the self-test step below executes them,
+    # so it must run the base-branch copies, not the PR head's.
+    - name: Save trusted hook config and gate scripts
       if: github.event_name == 'pull_request_target'
-      run: cp .pre-commit-config.yaml "$RUNNER_TEMP/trusted-pre-commit-config.yaml"
-    - uses: actions/checkout@v4
+      run: |
+        cp .pre-commit-config.yaml "$RUNNER_TEMP/trusted-pre-commit-config.yaml"
+        cp -a .github/scripts "$RUNNER_TEMP/trusted-scripts"
+        echo "GATE_SCRIPTS_DIR=$RUNNER_TEMP/trusted-scripts" >> "$GITHUB_ENV"
+    # allow-unsafe-pr-checkout acknowledges checkout's pull_request_target
+    # guard: the head is data for the trusted hooks to lint; nothing from it
+    # is executed (config and gate scripts are pinned to the base branch
+    # above) and credentials are not persisted. SHA-pinned to v4.4.0 because
+    # actionlint's action schema does not know the new input yet.
+    - uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4.4.0
       if: github.event_name == 'pull_request_target'
       with:
         ref: ${{ github.event.pull_request.head.sha }}
         persist-credentials: false
+        allow-unsafe-pr-checkout: true
     - name: Restore trusted hook config
       if: github.event_name == 'pull_request_target'
       run: cp "$RUNNER_TEMP/trusted-pre-commit-config.yaml" .pre-commit-config.yaml
@@ -48,5 +59,7 @@ jobs:
       with:
         extra_args: --all-files --hook-stage manual
     # After pre-commit so a self-test failure cannot mask lint failures.
+    # GATE_SCRIPTS_DIR points at the base-branch copy on fork PRs (set above);
+    # push / workflow_call runs use the checked-out tree directly.
     - name: Full-suite gate self-test
-      run: bash .github/scripts/test_gate_full_suite.sh
+      run: bash "${GATE_SCRIPTS_DIR:-.github/scripts}/test_gate_full_suite.sh"

.github/workflows/ci-scheduled-ssim.yml

@@ -0,0 +1,44 @@
+name: Scheduled Full SSIM
+
+on:
+  schedule:
+    - cron: "0 5 * * 0"
+  workflow_dispatch:
+
+permissions:
+  contents: read
+
+jobs:
+  trigger:
+    if: github.repository == 'hao-ai-lab/FastVideo'
+    runs-on: ubuntu-latest
+    steps:
+      - name: Trigger weekly full SSIM on Slinky Slurm
+        env:
+          BUILDKITE_API_TOKEN: ${{ secrets.BUILDKITE_API_TOKEN }}
+          SOURCE_SHA: ${{ github.sha }}
+          SOURCE_BRANCH: ${{ github.event.repository.default_branch }}
+          BK_ORG: ${{ vars.BUILDKITE_ORG_SLUG }}
+          BK_PIPELINE: ${{ vars.BUILDKITE_PIPELINE_SLUG }}
+        run: |
+          set -euo pipefail
+          curl -sS --fail-with-body -X POST \
+            "https://api.buildkite.com/v2/organizations/${BK_ORG}/pipelines/${BK_PIPELINE}/builds" \
+            -H "Authorization: Bearer $BUILDKITE_API_TOKEN" \
+            -H "Content-Type: application/json" \
+            --data-raw "$(jq -n \
+              --arg commit "$SOURCE_SHA" \
+              --arg branch "$SOURCE_BRANCH" \
+              '{
+                commit: $commit,
+                branch: $branch,
+                message: "Weekly full SSIM on Slinky Slurm",
+                ignore_pipeline_branch_filters: true,
+                env: {
+                  TEST_SCOPE: "scheduled",
+                  FULL_SUITE: "false",
+                  TEST_TYPE: "ssim",
+                  PR_NUMBER: "false",
+                  PR_TITLE: "Scheduled full SSIM"
+                }
+              }')"

.github/workflows/ci-slash-commands.yml

@@ -33,7 +33,6 @@ jobs:
             core.setOutput('has_write', String(hasWrite));
 
       - name: Add ready label and react
-        id: label
         if: steps.perm.outputs.has_write == 'true'
         uses: actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea # v7.0.1
         with:
@@ -48,47 +47,6 @@ jobs:
               comment_id: context.payload.comment.id,
               content: 'rocket',
             });
-            const { data: pr } = await github.rest.pulls.get({ owner, repo, pull_number: prNumber });
-            core.setOutput('pr_sha', pr.head.sha);
-            core.setOutput('pr_branch', pr.head.ref);
-            core.setOutput('pr_number', String(prNumber));
-            core.setOutput('pr_title', pr.title);
-
-      - name: Trigger Full Suite
-        if: steps.perm.outputs.has_write == 'true'
-        env:
-          BUILDKITE_API_TOKEN: ${{ secrets.BUILDKITE_API_TOKEN }}
-          PR_SHA: ${{ steps.label.outputs.pr_sha }}
-          PR_BRANCH: ${{ steps.label.outputs.pr_branch }}
-          PR_NUMBER: ${{ steps.label.outputs.pr_number }}
-          PR_TITLE: ${{ steps.label.outputs.pr_title }}
-          BK_ORG: ${{ vars.BUILDKITE_ORG_SLUG }}
-          BK_PIPELINE: ${{ vars.BUILDKITE_PIPELINE_SLUG }}
-        run: |
-          curl -sS --fail-with-body -X POST \
-            "https://api.buildkite.com/v2/organizations/${BK_ORG}/pipelines/${BK_PIPELINE}/builds" \
-            -H "Authorization: Bearer $BUILDKITE_API_TOKEN" \
-            -H "Content-Type: application/json" \
-            --data-raw "$(jq -n \
-              --arg commit "$PR_SHA" \
-              --arg branch "$PR_BRANCH" \
-              --arg message "Full Suite for PR #${PR_NUMBER} (via /merge)" \
-              --arg pr_title "$PR_TITLE" \
-              --argjson pr_id "$PR_NUMBER" \
-              '{
-                commit: $commit,
-                branch: $branch,
-                message: $message,
-                ignore_pipeline_branch_filters: true,
-                pull_request_id: $pr_id,
-                pull_request_base_branch: "main",
-                env: {
-                    TEST_SCOPE: "full",
-                    FULL_SUITE: "true",
-                    PR_NUMBER: ($pr_id | tostring),
-                    PR_TITLE: $pr_title
-                  }
-                }')"
 
   parse-command:
     if: >-
@@ -129,7 +87,7 @@ jobs:
           set -euo pipefail
           TEST_NAME=$(echo "$COMMENT" | grep -oP '(?<=/test\s)\S+' | head -1 || true)
 
-          VALID="encoder vae transformer kernel unit dreamverse ssim training lora-inference lora-training lora-extraction distillation self-forcing vsa vmoba performance api train-framework eval full fastcheck pre-commit"
+          VALID="encoder vae transformer kernel unit dreamverse ssim golden-gate training lora-inference lora-training lora-extraction distillation self-forcing vsa vmoba performance api train-framework eval unit-ci kernel-ci dreamverse-ci ssim-ci golden-gate-ci encoder-ci vae-ci transformer-ci lora-inference-ci lora-training-ci lora-extraction-ci training-ci distillation-ci self-forcing-ci vsa-ci vmoba-ci performance-ci api-ci train-framework-ci eval-ci full fastcheck pre-commit"
           if [ -z "$TEST_NAME" ] || ! echo "$VALID" | grep -qw "$TEST_NAME"; then
             echo "Unknown test: '$TEST_NAME'. Valid: $VALID"
             exit 1
@@ -137,8 +95,18 @@ jobs:
 
           declare -A MAP=(
             [encoder]=encoder [vae]=vae [transformer]=transformer
-            [kernel]=kernel_tests [unit]=unit_test [dreamverse]=dreamverse_app
-            [ssim]=ssim [training]=training
+            [kernel]=kernel_tests [unit]=unit_test [unit-ci]=unit_test_ci
+            [kernel-ci]=kernel_tests_ci [dreamverse-ci]=dreamverse_app_ci
+            [ssim-ci]=ssim_ci [vmoba-ci]=inference_vmoba_ci
+            [golden-gate-ci]=golden_gate_ci [training-ci]=training_ci
+            [encoder-ci]=encoder_ci [vae-ci]=vae_ci [transformer-ci]=transformer_ci
+            [lora-inference-ci]=inference_lora_ci [lora-training-ci]=training_lora_ci
+            [lora-extraction-ci]=lora_extraction_ci [distillation-ci]=distillation_dmd_ci
+            [self-forcing-ci]=self_forcing_ci [vsa-ci]=training_vsa_ci
+            [performance-ci]=performance_ci [api-ci]=api_server_ci
+            [train-framework-ci]=train_framework_ci [eval-ci]=eval_ci
+            [dreamverse]=dreamverse_app
+            [ssim]=ssim [golden-gate]=golden_gate [training]=training
             [lora-inference]=inference_lora [lora-training]=training_lora
             [lora-extraction]=lora_extraction
             [distillation]=distillation_dmd [self-forcing]=self_forcing

.github/workflows/ci-trigger-full-suite.yml

@@ -1,4 +1,4 @@
-name: Trigger Full Suite
+name: Trigger Merge Gate
 
 on:
   pull_request_target:
@@ -10,7 +10,7 @@ permissions:
   actions: read
 
 concurrency:
-  group: full-suite-${{ github.event.pull_request.number }}
+  group: merge-gate-${{ github.event.pull_request.number }}
   cancel-in-progress: false
 
 jobs:
@@ -34,29 +34,72 @@ jobs:
             });
             const hasReady = pr.labels.some(l => l.name === 'ready');
             core.setOutput('has_ready', String(hasReady));
-            if (!hasReady) core.info('No ready label — skipping Full Suite trigger.');
+            core.setOutput('changed_files', String(pr.changed_files));
+            if (!hasReady) core.info('No ready label — skipping merge-gate trigger.');
 
       - name: Cancel previous Buildkite builds
         if: steps.check.outputs.has_ready == 'true'
         env:
           BUILDKITE_API_TOKEN: ${{ secrets.BUILDKITE_API_TOKEN }}
           PR_BRANCH: ${{ github.event.pull_request.head.ref }}
+          PR_NUMBER: ${{ github.event.pull_request.number }}
         run: |
-          # Find running builds for this branch with TEST_SCOPE=full and cancel them
-          builds=$(curl -sS -H "Authorization: Bearer $BUILDKITE_API_TOKEN" \
-            "https://api.buildkite.com/v2/organizations/${{ vars.BUILDKITE_ORG_SLUG }}/pipelines/${{ vars.BUILDKITE_PIPELINE_SLUG }}/builds?branch=${PR_BRANCH}&state=running,scheduled" \
-            | jq -r '.[] | select(try (.env.TEST_SCOPE == "full") catch false) | .number')
+          # Match both branch and PR number: forks can reuse the same branch name.
+          builds=$(curl -sS --get -H "Authorization: Bearer $BUILDKITE_API_TOKEN" \
+            --data-urlencode "branch=$PR_BRANCH" \
+            --data-urlencode "state=running,scheduled" \
+            "https://api.buildkite.com/v2/organizations/${{ vars.BUILDKITE_ORG_SLUG }}/pipelines/${{ vars.BUILDKITE_PIPELINE_SLUG }}/builds" \
+            | jq -r --arg pr_number "$PR_NUMBER" \
+              '.[] | select((.env.TEST_SCOPE? == "merge") and (.env.PR_NUMBER? == $pr_number)) | .number')
           for build_num in $builds; do
             echo "Cancelling Buildkite build #$build_num"
             curl -sS -X PUT -H "Authorization: Bearer $BUILDKITE_API_TOKEN" \
               "https://api.buildkite.com/v2/organizations/${{ vars.BUILDKITE_ORG_SLUG }}/pipelines/${{ vars.BUILDKITE_PIPELINE_SLUG }}/builds/${build_num}/cancel"
           done
 
-      # Checks out the BASE branch (default for pull_request_target), so PR
-      # authors cannot tamper with the gate script.
-      - name: Checkout gate script
+      # Check out the immutable BASE SHA: pull_request_target must never run a
+      # planner or gate script from the untrusted PR head.
+      - name: Checkout trusted merge planner
         if: steps.check.outputs.has_ready == 'true'
         uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
+        with:
+          ref: ${{ github.event.pull_request.base.sha }}
+          persist-credentials: false
+
+      - name: Collect changed paths
+        if: steps.check.outputs.has_ready == 'true'
+        env:
+          GH_TOKEN: ${{ github.token }}
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          EXPECTED_CHANGED_FILES: ${{ steps.check.outputs.changed_files }}
+        run: |
+          set -euo pipefail
+          changed_json="$RUNNER_TEMP/merge-changed-files.json"
+          changed_paths="$RUNNER_TEMP/merge-changed-paths.txt"
+          if gh api --paginate --slurp \
+              "repos/${GITHUB_REPOSITORY}/pulls/${PR_NUMBER}/files?per_page=100" \
+              > "$changed_json"; then
+            observed=$(jq '[.[][] | .filename] | unique | length' "$changed_json")
+            if [ "$observed" = "$EXPECTED_CHANGED_FILES" ]; then
+              jq -r '.[][] | .filename, (.previous_filename // empty)' "$changed_json" \
+                | sort -u > "$changed_paths"
+            else
+              echo "::warning::Changed-file API returned $observed of $EXPECTED_CHANGED_FILES paths; selecting all merge lanes."
+              echo '__FASTVIDEO_CI_PLAN_ALL__' > "$changed_paths"
+            fi
+          else
+            echo "::warning::Changed-file API failed; selecting all merge lanes."
+            echo '__FASTVIDEO_CI_PLAN_ALL__' > "$changed_paths"
+          fi
+
+      - name: Select minimal merge tests
+        id: plan
+        if: steps.check.outputs.has_ready == 'true'
+        run: |
+          python3 .github/scripts/plan_merge_ci.py \
+            --paths-file "$RUNNER_TEMP/merge-changed-paths.txt" \
+            --github-output "$GITHUB_OUTPUT" \
+            --summary-file "$GITHUB_STEP_SUMMARY"
 
       - name: Wait for pre-commit and docs build
         if: steps.check.outputs.has_ready == 'true'
@@ -66,7 +109,7 @@ jobs:
           PR_NUMBER: ${{ github.event.pull_request.number }}
         run: bash .github/scripts/gate_full_suite.sh
 
-      - name: Trigger Buildkite Full Suite
+      - name: Trigger Buildkite merge gate
         if: steps.check.outputs.has_ready == 'true'
         env:
           BUILDKITE_API_TOKEN: ${{ secrets.BUILDKITE_API_TOKEN }}
@@ -76,6 +119,10 @@ jobs:
           PR_TITLE: ${{ github.event.pull_request.title }}
           BK_ORG: ${{ vars.BUILDKITE_ORG_SLUG }}
           BK_PIPELINE: ${{ vars.BUILDKITE_PIPELINE_SLUG }}
+          MERGE_TEST_PLAN: ${{ steps.plan.outputs.merge_test_plan }}
+          MERGE_GOLDEN_TESTS: ${{ steps.plan.outputs.merge_golden_tests }}
+          MERGE_SSIM_TESTS: ${{ steps.plan.outputs.merge_ssim_tests }}
+          MERGE_PLAN_LABEL: ${{ steps.plan.outputs.merge_plan_label }}
         run: |
           curl -sS --fail-with-body -X POST \
             "https://api.buildkite.com/v2/organizations/${BK_ORG}/pipelines/${BK_PIPELINE}/builds" \
@@ -84,8 +131,11 @@ jobs:
             --data-raw "$(jq -n \
               --arg commit "$PR_SHA" \
               --arg branch "$PR_BRANCH" \
-              --arg message "Full Suite for PR #${PR_NUMBER}" \
+              --arg message "Merge gate [${MERGE_PLAN_LABEL}] for PR #${PR_NUMBER}" \
               --arg pr_title "$PR_TITLE" \
+              --arg merge_test_plan "$MERGE_TEST_PLAN" \
+              --arg merge_golden_tests "$MERGE_GOLDEN_TESTS" \
+              --arg merge_ssim_tests "$MERGE_SSIM_TESTS" \
               --argjson pr_id "$PR_NUMBER" \
               '{
                 commit: $commit,
@@ -95,8 +145,11 @@ jobs:
                 pull_request_id: $pr_id,
                 pull_request_base_branch: "main",
                 env: {
-                  TEST_SCOPE: "full",
+                  TEST_SCOPE: "merge",
                   FULL_SUITE: "true",
+                  MERGE_TEST_PLAN: $merge_test_plan,
+                  MERGE_GOLDEN_TESTS: $merge_golden_tests,
+                  MERGE_SSIM_TESTS: $merge_ssim_tests,
                   PR_NUMBER: ($pr_id | tostring),
                   PR_TITLE: $pr_title
                 }

.github/workflows/community-welcome.yml

@@ -38,17 +38,17 @@ jobs:
 
             **How our CI works:**
 
-            PRs run a two-tier CI system:
+            PRs run a three-tier CI system:
             1. **Pre-commit** — formatting (yapf), linting (ruff), type checking (mypy). Runs immediately on every PR.
-            2. **Fastcheck** — core GPU tests (encoders, VAEs, transformers, kernels, unit tests). Runs automatically via Buildkite on relevant file changes (~10-15 min).
-            3. **Full Suite** — integration tests, training pipelines, SSIM regression. Runs only when a reviewer adds the `ready` label.
+            2. **Fastcheck** — six core GPU lanes run automatically via Buildkite (~10-15 min).
+            3. **Merge gate** — a reviewer adds `ready`; changed paths select only the relevant integration, training, golden, or SSIM coverage.
 
             **Before your PR is reviewed:**
             - [ ] `pre-commit run --all-files` passes locally
             - [ ] You've added or updated tests for your changes
             - [ ] The PR description explains what and why
 
-            If pre-commit fails, a bot comment will explain how to fix it. Fastcheck and Full Suite results appear in the Checks section below.
+            If pre-commit fails, a bot comment will explain how to fix it. Fastcheck and merge-gate results appear in the Checks section below.
 
             **Useful links:**
             - [Contributing Guide](https://hao-ai-lab.github.io/FastVideo/contributing/overview/)

.github/workflows/infra-build-image.yml

@@ -13,16 +13,28 @@ on:
         required: false
         default: false
         type: boolean
-  # Auto-rebuild the CUDA images when their Dockerfile changes on main. The CUDA
-  # matrix is the only lane that builds from docker/Dockerfile, so a path-scoped
-  # push trigger is a sufficient change detector on its own -- no separate
-  # detect-changes/paths-filter job is needed now that there is a single
-  # in-scope Dockerfile. Dreamverse (apps/dreamverse/docker/Dockerfile) and the
-  # rocm Dockerfile stay manual-dispatch only.
+      build_ci_runner_image:
+        description: 'Build the ARM64 CUDA 13 CI runner image (sm_100)'
+        required: false
+        default: false
+        type: boolean
+  # Auto-rebuild the CUDA images when a repository-controlled image input
+  # changes on main. This includes the trusted SM89 kernel artifact's source,
+  # metadata/key helper, ABI dependency metadata, and build orchestration.
+  # Dreamverse (apps/dreamverse/docker/Dockerfile) and the ROCm Dockerfile stay
+  # manual-dispatch only.
   push:
     branches: [main]
     paths:
+      - '.dockerignore'
+      - '.github/workflows/_template-build-image.yml'
+      - '.github/workflows/infra-build-image.yml'
+      - '.gitmodules'
       - 'docker/Dockerfile'
+      - 'docker/uv-excludes'
+      - 'fastvideo-kernel/**'
+      - 'fastvideo/tests/modal/kernel_build_cache.py'
+      - 'pyproject.toml'
 
 
 permissions:
@@ -50,7 +62,7 @@ jobs:
   # 2.8.3 comes from the architecture-specific prebuilt releases.
   build-cuda-images:
     # Runs on a manual dispatch when build_cuda_matrix is set, or automatically
-    # on a push that changed docker/Dockerfile (inputs are null on push). The
+    # on an in-scope main push (inputs are null on push). The
     # repository guard keeps fork syncs from auto-building; manual dispatch
     # still works in forks.
     if: ${{ (github.event_name == 'push' && github.repository == 'hao-ai-lab/FastVideo') || github.event.inputs.build_cuda_matrix == 'true' }}
@@ -191,6 +203,29 @@ jobs:
           docker buildx imagetools create "${TAG_ARGS[@]}" "${IMAGE_REFS[@]}"
           docker buildx imagetools inspect "${TAGS[0]}"
 
+  # The CI runner is ARM64 like DGX Spark, but targets sm_100a rather than sm_121.
+  # The architecture-specific target includes the GB200 VSA CUDA extensions.
+  # Publish a single-architecture variant so the self-hosted CI runner can reuse
+  # the exact prebuilt kernel instead of compiling it in every job.
+  build-ci-runner-image:
+    if: ${{ (github.event_name == 'push' && github.repository == 'hao-ai-lab/FastVideo') || github.event.inputs.build_ci_runner_image == 'true' }}
+    uses: ./.github/workflows/_template-build-image.yml
+    with:
+      python_version: '3.12'
+      dockerfile_path: docker/Dockerfile
+      tag_suffix: py3.12-cuda13.0.0-sm100
+      runner: ubuntu-24.04-arm
+      architecture: arm64
+      build_args: |
+        PYTHON_VERSION=3.12
+        CUDA_VERSION=13.0.0
+        UV_TORCH_BACKEND=cu130
+        TORCH_CUDA_ARCH_LIST=10.0a
+        CMAKE_BUILD_PARALLEL_LEVEL=1
+        FLASH_ATTN_WHEEL_TAG=cu130torch2.12
+        FLASH_ATTN_WHEEL_RELEASE_ARM64=https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.9.22
+    secrets: inherit
+
   # Dreamverse matrix: {backend, UI} x {12.6.3, 13.0.0}, Python 3.12. Torch backend
   # matches the base CUDA (cu126 / cu130). Keep these images amd64-only until the
   # required FA4 dependency stack is available and validated on arm64.

.github/workflows/infra-docs.yml

@@ -6,6 +6,7 @@ on:
     paths:
       - 'docs/**'
       - 'examples/**'
+      - 'scripts/inference/**'
       - 'mkdocs.yml'
       - 'requirements-mkdocs.in'
       - 'requirements-mkdocs.txt'
@@ -16,6 +17,7 @@ on:
     paths:
       - 'docs/**'
       - 'examples/**'
+      - 'scripts/inference/**'
       - 'mkdocs.yml'
       - 'requirements-mkdocs.in'
       - 'requirements-mkdocs.txt'

.github/workflows/publish-kernel.yml

@@ -62,8 +62,9 @@ jobs:
               cuda-version: '13.0.0'
               torch-cuda-short: 'cu130'
           platform:
-            # x86_64 builds the full cu126 + cu130 set (cu130 ships the consumer
-            # Blackwell sm_120a FP4 kernels).
+            # x86_64 builds the full cu126 + cu130 set. cu130 ships the
+            # data-center Blackwell sm_100a/sm_103a VSA and consumer sm_120a FP4
+            # kernels.
             - os: ubuntu-22.04
               arch: x86_64
               wheel-plat: manylinux_2_35_x86_64
@@ -124,7 +125,7 @@ jobs:
       - name: Install dependencies (GCC, Clang, CUDA Paths, Git)
         run: |
           sudo apt update
-          sudo apt install -y git patchelf gcc-11 g++-11 clang-11
+          sudo apt install -y git gcc-11 g++-11 clang-11
           sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
 
           # Allow Git to Access Safe Directory
@@ -168,17 +169,18 @@ jobs:
           #     covers sm_120a; turbodiffusion covers sm_100a+sm_120a. The sm_100 FP4
           #     forward is the FA4 CuTe DSL path in the fastvideo package (PR #1221),
           #     JIT-compiled at runtime — not built into this wheel.
-          #   * x86_64 cu130 = Hopper TK + consumer Blackwell sm_120a FP4.
+          #   * x86_64 cu130 = Hopper TK + data-center Blackwell sm_100a/sm_103a VSA
+          #     + consumer Blackwell sm_120a FP4.
           #   * x86_64 cu126 = Hopper TK only (older drivers; CUDA < 12.8 has no FP4).
           # The per-arch split in CMakeLists pins the FP4 targets to sm_120a and builds
           # the main extension for the full arch list. CMAKE_BUILD_PARALLEL_LEVEL caps
           # Ninja so heavy CUTLASS/TK template TUs don't OOM the 16 GB runner (exit 143).
           if [ "${{ matrix.platform.arch }}" = "aarch64" ]; then
-            export TORCH_CUDA_ARCH_LIST="10.0a;12.0a"
+            export TORCH_CUDA_ARCH_LIST="10.0a;10.3a;12.0a"
             export CMAKE_ARGS="${CMAKE_ARGS:-} -DFASTVIDEO_KERNEL_BUILD_TK=OFF -DFASTVIDEO_KERNEL_BUILD_ATTN_QAT_INFER=ON"
             export CMAKE_BUILD_PARALLEL_LEVEL=1
           elif [ "${{ matrix.torch-cuda.torch-cuda-short }}" = "cu130" ]; then
-            export TORCH_CUDA_ARCH_LIST="9.0a;12.0a"
+            export TORCH_CUDA_ARCH_LIST="9.0a;10.0a;10.3a;12.0a"
             export CMAKE_ARGS="${CMAKE_ARGS:-} -DFASTVIDEO_KERNEL_BUILD_TK=ON -DFASTVIDEO_KERNEL_BUILD_ATTN_QAT_INFER=ON -DCMAKE_CUDA_ARCHITECTURES=90a"
             # A single FP4 TU (attn_qat_infer) can use ~8-12 GB on its own, so serialize.
             export CMAKE_BUILD_PARALLEL_LEVEL=1
@@ -194,7 +196,11 @@ jobs:
           python -m build --wheel --outdir dist
 
           # Fix the wheel to be manylinux compliant
-          uv pip install --system auditwheel
+          # Ubuntu 22.04 ships patchelf 0.14.3, while current auditwheel
+          # requires at least 0.14.5. Use the stable PyPI binary on both
+          # x86_64 and aarch64 release runners.
+          uv pip install --system auditwheel patchelf==0.17.2.4
+          patchelf --version
           # Point auditwheel at torch libs, but do not vendor them into the wheel.
           TORCH_LIB_DIR=$(python - <<'PY'
           import os
@@ -211,7 +217,8 @@ jobs:
             --exclude libtorch.so \
             --exclude libc10.so \
             --exclude libc10_cuda.so \
-            --exclude libtorch_python.so
+            --exclude libtorch_python.so \
+            --exclude libnccl.so.2
           # Move fixed wheels back to dist for upload consistency
           rm dist/*.whl
           mv fixed_dist/*.whl dist/

.gitignore

@@ -6,6 +6,7 @@ results/
 wandb/
 *.ipynb
 *.jpg
+!examples/datasets/lingbotworld2/image.jpg
 *.safetensors
 *.mp4
 *.png
@@ -22,6 +23,7 @@ Miniconda3-latest-Linux-x86_64.sh
 *validation/
 data/
 outputs/
+outputs_audio/
 outputs_video
 checkpoints/
 sbatch.sh
@@ -34,14 +36,10 @@ env
 *.log
 weights/
 logs/
+/Z-Image/
 official_weights/
 converted_weights/
 
-# Cosmos3 local parity assets (symlinked from main worktree)
-/official_weights/
-/converted_weights/
-/cosmos-framework
-
 # SSIM test outputs
 fastvideo/tests/ssim/generated_videos/
 **/.cache/**
@@ -57,6 +55,8 @@ eggs/
 
 # MkDocs documentation
 site/
+docs/assets/cookbook-serving.json
+examples/serving/clients/node_modules/
 docs/getting_started/examples/
 docs/examples/
 docs/inference/examples/
@@ -78,6 +78,7 @@ docs/distillation/examples/
 *.pkl
 
 # Reference videos (negations must come after the catch-all on line below)
+!fastvideo/tests/nightly/reference_video_*.mp4
 
 # Static images
 !docs/assets/images/**/*.png
@@ -134,6 +135,9 @@ fastvideo/tests/ssim/reference_videos/**
 !fastvideo/tests/ssim/reference_videos/**/*.mp4
 !fastvideo/tests/ssim/reference_videos/**/*.png
 
+# Local H3 MLX kernel / exactness benches (JSON, logs, frames, videos)
+.kernel_bench/
+
 # Editor logs and local Python version pins (accidentally committed)
 *.nvimlog
 .nvimlog

.pre-commit-config.yaml

@@ -9,7 +9,7 @@ exclude: |
     tests/.*|
     scripts/.*|
     fastvideo/dataset/.*|
-    fastvideo/models/.*|
+    fastvideo/models/(?!wan/(config|vae_config|pipeline_config|definition|__init__)\.py$).*|
     ^apps/dreamverse/web/.*|
     examples/.*|
     \.agents/.*|
@@ -22,6 +22,7 @@ repos:
   hooks:
   - id: yapf
     args: [--in-place, --verbose]
+    language_version: python3.12
     additional_dependencies: [toml] # TODO: Remove when yapf is upgraded
 - repo: https://github.com/astral-sh/ruff-pre-commit
   rev: v0.11.12

AGENTS.md

@@ -66,14 +66,18 @@ Local guidance lives next to the code. Read the in-scope file before editing:
 | `fastvideo/AGENTS.md` | Core package map, public API, registry-driven model dispatch |
 | `fastvideo/configs/AGENTS.md` | Arch + pipeline config dataclasses, `param_names_mapping` |
 | `fastvideo/models/AGENTS.md` | DiT / VAE / encoder / scheduler / loader layout (pre-commit excluded) |
+| `fastvideo/models/wan/AGENTS.md` | Wan family-local transformers, VAE, configs, and the SP sharding invariant |
 | `fastvideo/layers/AGENTS.md` | Tensor-parallel linear/attention layer rules for ports |
 | `fastvideo/attention/AGENTS.md` | Backend registry + env-var override |
 | `fastvideo/pipelines/AGENTS.md` | Stage ABC, `basic/<model>/`, `preprocess/`, presets |
+| `fastvideo/pipelines/basic/wan/AGENTS.md` | Wan sampling stages, first-frame conditioning, DMD/causal boundaries |
+| `fastvideo/pipelines/basic/magi_human/AGENTS.md` | MagiHuman umbrella repo, lazy-loaded components, packing invariants |
 | `fastvideo/training/AGENTS.md` | Legacy monolithic pipelines (frozen for existing models) |
 | `fastvideo/train/AGENTS.md` | New modular trainer (methods × models × callbacks, YAML) |
 | `fastvideo/tests/AGENTS.md` | Test taxonomy, conftest, pre-commit-excluded path |
 | `fastvideo/tests/ssim/AGENTS.md` | GPU SSIM regression authoring + reference video sync |
 | `scripts/checkpoint_conversion/AGENTS.md` | Adding a converter for a new HF/official checkpoint |
+| `apps/dreamverse/AGENTS.md` | DreamVerse app structure and conventions |
 
 ## Critical: Two Training Stacks Coexist
 

README.md

@@ -3,13 +3,17 @@
 </div>
 
 <p align="center">
-     | <a href="https://hao-ai-lab.github.io/FastVideo"><b>Documentation</b></a> | <a href="https://hao-ai-lab.github.io/FastVideo/inference/inference_quick_start/"><b> Quick Start</b></a> | <a href="https://github.com/hao-ai-lab/FastVideo/discussions/982"  target="_blank"><b>Weekly Dev Meeting</b></a>  | 🟣💬 <a href="https://join.slack.com/t/fastvideo/shared_invite/zt-3f4lao1uq-u~Ipx6Lt4J27AlD2y~IdLQ" target="_blank"> <b>Slack</b> </a> |  🟣💬 <a href="https://github.com/hao-ai-lab/FastVideo/discussions/1097" target="_blank"> <b> WeChat </b> </a> |
+     | <a href="https://hao-ai-lab.github.io/FastVideo"><b>Documentation</b></a> | <a href="https://haoailab.com/FastVideo/cookbook/"><b>Cookbook</b></a> | <a href="https://hao-ai-lab.github.io/FastVideo/inference/inference_quick_start/"><b> Quick Start</b></a> | <a href="https://github.com/hao-ai-lab/FastVideo/discussions/982"  target="_blank"><b>Weekly Dev Meeting</b></a>  | 🟣💬 <a href="https://join.slack.com/t/fastvideo/shared_invite/zt-3f4lao1uq-u~Ipx6Lt4J27AlD2y~IdLQ" target="_blank"> <b>Slack</b> </a> |  🟣💬 <a href="https://github.com/hao-ai-lab/FastVideo/discussions/1097" target="_blank"> <b> WeChat </b> </a> |
 </p>
 
 **FastVideo is a unified post-training and real-time inference framework for accelerated video generation.**
 
 ## NEWS
-- `2026/06/23`: Release FastWan-QAD: 5s of Video generated in 1.8s E2E. [FastWan-QAD models](https://huggingface.co/FastVideo/FastWan-QAD-FP8-1.3B), check out the [Blog](https://haoailab.com/blogs/fastwan-qad/).
+- `2026/09/15`: Release [FastH3 8-Step V2](https://huggingface.co/FastVideo/FastVideo-FastH3-8-Step-V2), an eight-forward data-free DMD2 checkpoint distilled from MiniMax-H3 with 80% Video Sparse Attention. Run it with `examples/inference/basic/basic_fasth3_8step.py` or the [FastH3 8-Step V2 recipe](https://haoailab.com/FastVideo/cookbook/minimax-h3/).
+- `2026/09/01`: FastH3 now runs locally on Apple Silicon through MLX and on NVIDIA DGX Spark through CUDA 13, including two-Spark inference. Follow the [FastH3 recipes](https://haoailab.com/FastVideo/cookbook/minimax-h3/) and read the [Blog](https://haoailab.com/blogs/fasth3-local/).
+- `2026/08/27`: [FastH3 Preview v1](https://haoailab.com/blogs/fasth3-preview/) is an open-weight 4-step sparse-distilled MiniMax-H3 model for synchronized video-and-audio generation, developed in collaboration with [Nuva Lab](https://nuvalab.ai/) and the [NVIDIA FastGen team](https://github.com/NVlabs/FastGen). Download the recommended [VSA / Data-Free weights](https://huggingface.co/FastVideo/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree), or see the [full FastH3 collection](https://huggingface.co/collections/FastVideo/fastvideo-fasth3).
+- `2026/08/19`: FastVideo now supports MLX on Apple Silicon with [FastMetal-QAD](https://huggingface.co/collections/FastVideo/fastmetal), a family of 1.3B, 5B, and 14B models optimized for Mac. Follow the [MLX install guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/mlx/) and read the [Blog](https://haoailab.com/blogs/fastmetal/).
+- `2026/06/23`: Release FastWan-QAD: 5s of Video generated in 1.8s E2E. See the [FastWan-QAD models](https://huggingface.co/FastVideo/FastWan-QAD-FP8-1.3B), [Attn-QAT training guide](https://haoailab.com/FastVideo/training/attn_qat/), and [blog](https://haoailab.com/blogs/fastwan-qad/).
 - `2026/03/17`: Release demo: Into the Dreamverse: Vibe Directing in FastVideo, check out the [Blog](https://haoailab.com/blogs/dreamverse/).
 - `2026/03/13`: Release demo: Create a 5s 1080p Video in 4.5s with FastVideo on a Single GPU, check out the [Blog](https://haoailab.com/blogs/fastvideo_realtime_1080p/).
 - `2025/11/19`: Release [CausalWan2.2 I2V A14B Preview](https://huggingface.co/FastVideo/CausalWan2.2-I2V-A14B-Preview-Diffusers) models, [Blog](https://hao-ai-lab.github.io/blogs/fastvideo_causalwan_preview/) and [Inference Code!](https://github.com/hao-ai-lab/FastVideo/blob/main/examples/inference/basic/basic_self_forcing_causal_wan2_2_i2v.py).
@@ -33,7 +37,7 @@ FastVideo has the following features:
   - [Sparse distillation](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/) to achieve >50x denoising speedup
   - Scalable training with FSDP2, sequence parallelism, and selective activation checkpointing.
   - Causal distillation through Self-Forcing
-  - See this [page](https://hao-ai-lab.github.io/FastVideo/training/overview/) for full list of supported models and recipes.
+  - See this [page](https://hao-ai-lab.github.io/FastVideo/training/overview/) for the supported training workflows, and the [support matrix](https://hao-ai-lab.github.io/FastVideo/inference/support_matrix/) for supported models.
 - State-of-the-art performance optimizations for inference
   - Sequence Parallelism for distributed inference
   - Multiple state-of-the-art attention backends
@@ -60,7 +64,12 @@ UV_TORCH_BACKEND=cu126 uv pip install fastvideo
 ```
 
 Use `UV_TORCH_BACKEND=cu130` on CUDA 13. Apple silicon users should follow the
-[MPS installation guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/mps/).
+[MLX install guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/mlx/).
+
+> **On an Apple Silicon Mac?** Install with `uv pip install -e '.[mlx]'` from
+> a clone, then pick a recipe in the
+> [cookbook](https://haoailab.com/FastVideo/cookbook/). See the
+> [MLX install guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/mlx/).
 
 Please see our [docs](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/) for more detailed installation instructions.
 
@@ -78,7 +87,7 @@ Install FastVideo (https://github.com/hao-ai-lab/FastVideo) into a fresh uv virt
    https://hao-ai-lab.github.io/FastVideo/getting_started/installation/):
      - NVIDIA GPU, x86_64                         -> docs/getting_started/installation/gpu.md
      - NVIDIA DGX Spark / GB10, aarch64, CUDA 13  -> docs/getting_started/installation/spark.md
-     - Apple Silicon, macOS                       -> docs/getting_started/installation/mps.md
+     - Apple Silicon, macOS                       -> docs/getting_started/installation/mlx.md
 3. Use uv for every step. If a command fails, debug it and tell me what you changed.
 4. Verify the result:
      python -c "import fastvideo, torch; print('cuda', torch.cuda.is_available())"

apps/dreamverse/README.md

@@ -97,13 +97,33 @@ dreamverse-server --port 8009
 dreamverse-mock-server --port 8009
 ```
 
+### Run Dreamverse with FastH3
+
+Select the VSA data-free FastH3 Preview profile when you start the backend:
+
+```bash
+DREAMVERSE_MODEL_ID=fast-h3 dreamverse-server --port 8009
+```
+
+The `fast-h3` profile uses four visible GPUs by default. It loads the `MiniMaxAI/MiniMax-H3` base checkpoint and the
+`vsa-datafree/adapter_model.safetensors` adapter from
+`FastVideo/FastVideo-FastH3-4-step-Preview-v1-LoRA`. Each request generates a 124-frame, 768×1344 video with
+synchronized audio and five sigma-grid points. Dreamverse uses the last frame of each segment as first-frame
+conditioning for the following segment.
+
+Set `CUDA_VISIBLE_DEVICES` when you need to choose the four physical GPUs:
+
+```bash
+CUDA_VISIBLE_DEVICES=0,1,2,3 DREAMVERSE_MODEL_ID=fast-h3 dreamverse-server --port 8009
+```
+
 > **Expect a slow first boot.** With `torch.compile` and startup warmup enabled
 > (the default), the backend compiles the segment 1 and segment 2 inference
 > paths before it reports ready — this can take **tens of minutes on a cold
 > cache**, regardless of how you deploy (local, server, Docker, or Modal).
 > `/healthz` responds as soon as the process is up; `/readyz` stays `503` until
-> warmup finishes. For a faster, uncompiled startup while testing, set
-> `FASTVIDEO_ENABLE_STARTUP_WARMUP=0` before starting the backend.
+> warmup finishes. To defer compilation until the first generated request while
+> testing, set `FASTVIDEO_ENABLE_STARTUP_WARMUP=0` before starting the backend.
 
 ## Frontend Setup
 
@@ -219,6 +239,7 @@ selection, and mock-server behavior:
 pytest apps/dreamverse/dreamverse/tests/test_config.py \
   apps/dreamverse/dreamverse/tests/test_entrypoints.py \
   apps/dreamverse/dreamverse/tests/test_gpu_pool.py \
+  apps/dreamverse/dreamverse/tests/test_minimax_h3_generation.py \
   apps/dreamverse/dreamverse/tests/test_mock_server.py -q
 ```
 

apps/dreamverse/arch.md

@@ -139,7 +139,18 @@ session.
 - startup warmup
 - user join/leave commands
 - `USER_STEP` execution for each segment
-- continuation state between segments
+- generation-command routing and stream-result delivery
+
+Model generation has a separate ownership boundary inside each GPU process:
+
+- `apps/dreamverse/dreamverse/generation_worker.py` selects the backend that the active model profile declares and owns
+  the backend lifecycle.
+- `apps/dreamverse/dreamverse/ltx2_generation.py` owns LTX-2 generator configuration, video and audio continuation, and
+  runtime LoRA application.
+- `apps/dreamverse/dreamverse/minimax_h3_generation.py` owns the VSA data-free FastH3 adapter, FastH3 generator and
+  request configuration, and last-frame continuation through MiniMax H3 first-frame conditioning.
+- `apps/dreamverse/dreamverse/generation_contracts.py` defines the decoded media and stream-trimming result that both
+  model backends return to `apps/dreamverse/dreamverse/gpu_pool.py`.
 
 `apps/dreamverse/dreamverse/prompt_enhancer.py` manages:
 

apps/dreamverse/dreamverse/benchmarks/benchmark_pipeline.py

@@ -1,6 +1,6 @@
 """Benchmark the LTX-2 generation pipeline driven by the dreamverse Python SDK path.
 
-Mirrors how ``apps/dreamverse/dreamverse/video_generation.py`` constructs
+Mirrors how ``apps/dreamverse/dreamverse/ltx2_generation.py`` constructs
 ``GeneratorConfig`` and calls ``VideoGenerator.generate()``, then
 captures per-stage timings via the ``FASTVIDEO_STAGE_LOGGING=1`` log
 hooks (same mechanism as ``FastVideo-internal/examples/inference/basic/

apps/dreamverse/dreamverse/config.py

@@ -1,5 +1,6 @@
 import os
 from pathlib import Path
+from typing import cast
 
 _REPO_ROOT = Path(__file__).resolve().parents[1]
 _SERVER_ROOT = Path(__file__).resolve().parent
@@ -55,16 +56,34 @@ FRONTEND_STATIC_DIR_CANDIDATES = _resolve_frontend_static_dir_candidates()
 MODEL_REGISTRY = {
     "fast-ltx2": {
         "name": "FastLTX2",
+        "generation_backend": "ltx2",
+        "default_sp_size": 1,
         "model_path": "FastVideo/LTX2-Distilled-Diffusers",
         "config_model_path": "FastVideo/LTX2-Distilled-Diffusers",
         "lora_repo": "FastVideo/LTX2-OmniNFT-LoRA",
     },
     "fast-ltx23": {
         "name": "FastLTX23",
+        "generation_backend": "ltx2",
+        "default_sp_size": 1,
         "model_path": "FastVideo/LTX-2.3-Distilled-Diffusers",
         "config_model_path": "FastVideo/LTX-2.3-Distilled-Diffusers",
         "lora_repo": "FastVideo/LTX-2.3-OmniNFT-LoRA",
     },
+    "fast-h3": {
+        "name": "FastH3",
+        "generation_backend": "minimax_h3",
+        "default_sp_size": 4,
+        "model_path": "MiniMaxAI/MiniMax-H3",
+        "adapter_repo": "FastVideo/FastVideo-FastH3-4-step-Preview-v1-LoRA",
+        "adapter_filename": "vsa-datafree/adapter_model.safetensors",
+        "attention_backend": "VIDEO_SPARSE_ATTN_H3",
+        "height": 768,
+        "width": 1344,
+        "num_frames": 124,
+        "num_inference_steps": 5,
+        "seed": 1000,
+    },
 }
 
 DEFAULT_MODEL_ID = "fast-ltx2"
@@ -171,7 +190,7 @@ def _optional_env(*names: str) -> str | None:
 DEVTOOLS_ENABLED = _env_bool("FASTVIDEO_ENABLE_DEVTOOLS", False)
 PROMPT_SAFETY_ENABLED = _env_bool("FASTVIDEO_ENABLE_PROMPT_SAFETY", False)
 DREAMVERSE_MAX_AUTOTUNE = _env_bool("DREAMVERSE_MAX_AUTOTUNE", True)
-DREAMVERSE_SP_SIZE = max(1, _env_int("DREAMVERSE_SP_SIZE", 1))
+DREAMVERSE_SP_SIZE = max(1, _env_int("DREAMVERSE_SP_SIZE", cast(int, MODEL_CONFIG["default_sp_size"])))
 
 DREAMVERSE_MODEL_PATH = (os.getenv("DREAMVERSE_MODEL_PATH", "").strip() or None)
 if DREAMVERSE_MODEL_PATH:
@@ -213,7 +232,7 @@ def _resolve_lora_spec(spec: str) -> str | None:
     if not spec:
         return None
     if spec.lower() == "omninft":
-        return MODEL_CONFIG.get("lora_repo")
+        return cast(str | None, MODEL_CONFIG.get("lora_repo"))
     if spec.lower() in AVAILABLE_LORAS:
         return AVAILABLE_LORAS[spec.lower()]["repo"]
     return spec

apps/dreamverse/dreamverse/generation_contracts.py

@@ -0,0 +1,46 @@
+"""Shared contract between DreamVerse generation backends and GPU workers."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any, Protocol
+
+
+@dataclass
+class StepResult:
+    """Decoded media and stream-trimming metadata for one DreamVerse segment."""
+
+    frames: list
+    audio: Any
+    audio_sample_rate: int | None
+    timings: dict[str, float]
+    head_trim_frames: int
+    head_trim_audio_frames: int
+
+
+class GenerationBackend(Protocol):
+    """Model-owned generation operations used by one GPU worker process."""
+
+    def initialize(self, model_config: dict | None = None) -> None:
+        ...
+
+    def shutdown(self) -> None:
+        ...
+
+    def clear_conditioning(self) -> None:
+        ...
+
+    def generate_step(
+        self,
+        prompt: str,
+        segment_idx: int,
+        image_path: str | None,
+        reset_conditioning: bool,
+    ) -> StepResult:
+        ...
+
+    def warmup(self, prompt: str) -> dict[str, float]:
+        ...
+
+    def apply_lora_stack(self, stack: list[tuple[str, float]]) -> tuple[str | None, str | None]:
+        ...

apps/dreamverse/dreamverse/generation_worker.py

@@ -0,0 +1,96 @@
+"""Select and own one model-specific generation backend per GPU process."""
+
+from __future__ import annotations
+
+from dreamverse.config import MODEL_CONFIG
+from dreamverse.generation_contracts import GenerationBackend, StepResult
+
+
+def _create_generation_backend(backend_name: str, gpu_id: int) -> GenerationBackend:
+    """Construct the backend that owns the selected model family's behavior."""
+    if backend_name == "ltx2":
+        from dreamverse.ltx2_generation import LTX2GenerationBackend
+
+        return LTX2GenerationBackend(gpu_id)
+    if backend_name == "minimax_h3":
+        from dreamverse.minimax_h3_generation import MiniMaxH3GenerationBackend
+
+        return MiniMaxH3GenerationBackend(gpu_id)
+    raise ValueError(f"Unsupported DreamVerse generation backend: {backend_name!r}")
+
+
+class VideoGenerationWorker:
+    """Delegate GPU lifecycle and generation calls to the active model backend."""
+
+    def __init__(self, gpu_id: int):
+        self.gpu_id = gpu_id
+        self.model_config: dict = dict(MODEL_CONFIG)
+        self.backend_name: str | None = None
+        self.backend: GenerationBackend | None = None
+
+    def initialize(self, model_config: dict | None = None) -> None:
+        """Load the requested model through its generation backend.
+
+        Model selection belongs here so the GPU process and streaming layers
+        use one stable media contract without importing model-specific code.
+        """
+        requested_model_config = dict(model_config) if model_config is not None else dict(self.model_config)
+        backend_name = requested_model_config.get("generation_backend")
+        if not isinstance(backend_name, str) or not backend_name:
+            raise ValueError("DreamVerse model configuration requires `generation_backend`.")
+
+        candidate_backend = self.backend
+        if candidate_backend is None or self.backend_name != backend_name:
+            if candidate_backend is not None:
+                candidate_backend.shutdown()
+            candidate_backend = _create_generation_backend(backend_name, self.gpu_id)
+
+        try:
+            candidate_backend.initialize(requested_model_config)
+        except Exception:
+            try:
+                candidate_backend.shutdown()
+            except Exception as shutdown_error:
+                print(f"[GPU {self.gpu_id}] Backend cleanup after initialization failure: {shutdown_error}")
+            self.backend = None
+            self.backend_name = None
+            raise
+
+        self.model_config = requested_model_config
+        self.backend = candidate_backend
+        self.backend_name = backend_name
+
+    def _require_backend(self) -> GenerationBackend:
+        """Return the initialized backend or fail before processing a command."""
+        if self.backend is None:
+            raise RuntimeError("Generation backend is not initialized.")
+        return self.backend
+
+    def shutdown(self) -> None:
+        """Release model resources owned by the selected backend."""
+        if self.backend is not None:
+            self.backend.shutdown()
+
+    def clear_conditioning(self) -> None:
+        self._require_backend().clear_conditioning()
+
+    def generate_step(
+        self,
+        prompt: str,
+        segment_idx: int,
+        image_path: str | None,
+        reset_conditioning: bool,
+    ) -> StepResult:
+        """Generate one segment through the selected model backend."""
+        return self._require_backend().generate_step(
+            prompt,
+            segment_idx,
+            image_path,
+            reset_conditioning,
+        )
+
+    def warmup(self, prompt: str) -> dict[str, float]:
+        return self._require_backend().warmup(prompt)
+
+    def apply_lora_stack(self, stack: list[tuple[str, float]]) -> tuple[str | None, str | None]:
+        return self._require_backend().apply_lora_stack(stack)

apps/dreamverse/dreamverse/gpu_pool.py

@@ -12,7 +12,7 @@ from enum import Enum
 from multiprocessing import Process, Queue
 
 from dreamverse.config import (
-    DEFAULT_MODEL_ID,
+    ACTIVE_MODEL_ID,
     DREAMVERSE_SP_SIZE,
     MODEL_REGISTRY,
     STARTUP_WARMUP_ENABLED,
@@ -54,7 +54,7 @@ from dreamverse.worker_ipc import (
 def _parse_requested_gpu_limit() -> int | None:
     raw_value = os.getenv("FASTVIDEO_GPU_COUNT", "").strip().lower()
     if not raw_value:
-        return 1
+        return DREAMVERSE_SP_SIZE
     if raw_value == "all":
         return None
     try:
@@ -164,12 +164,12 @@ def gpu_worker_process(
     os.environ["CUDA_VISIBLE_DEVICES"] = cuda_device
     os.environ["FASTVIDEO_ATTENTION_BACKEND"] = "FLASH_ATTN"
 
-    from dreamverse.video_generation import VideoGenerationWorker
+    from dreamverse.generation_worker import VideoGenerationWorker
 
     worker = VideoGenerationWorker(gpu_id)
 
     def event_loop(first_cmd: Command = None):
-        """Blocking event loop for LTX2; dispatches user commands."""
+        """Block on generation commands after the model is initialized."""
         print(f"[GPU {gpu_id}] Entering event loop")
 
         def handle_command(cmd: Command):
@@ -435,7 +435,7 @@ class GPUSlot:
         self._response_reader_task: asyncio.Task | None = None
         self._active: bool = False
         self._reader_lock: asyncio.Lock | None = None
-        self.current_model_id: str = DEFAULT_MODEL_ID
+        self.current_model_id: str | None = ACTIVE_MODEL_ID
         self.shared_stream_buffer = None
         self.shared_stream_buffer_size = SHARED_STREAM_BUFFER_BYTES
 
@@ -690,7 +690,7 @@ class GPUSlot:
     async def join_user(self, user_id: str, model_id: str = None) -> JoinAck:
         """Add a user to this GPU."""
         if model_id is None:
-            model_id = DEFAULT_MODEL_ID
+            model_id = ACTIVE_MODEL_ID
 
         # Reload model if a different one is requested
         if model_id != self.current_model_id and model_id in MODEL_REGISTRY:
@@ -705,16 +705,23 @@ class GPUSlot:
             self.connected_users.clear()
 
             model_config = MODEL_REGISTRY[model_id]
-            reload_response = await self._send_command(Command(CommandType.RELOAD_MODEL,
-                                                               payload=ReloadModelPayload(model_config=model_config),
-                                                               user_id="__reload__"),
-                                                       timeout=600.0)
+            try:
+                reload_response = await self._send_command(Command(
+                    CommandType.RELOAD_MODEL,
+                    payload=ReloadModelPayload(model_config=model_config),
+                    user_id="__reload__"),
+                                                           timeout=600.0)
+            except Exception:
+                self.current_model_id = None
+                raise
             match reload_response:
                 case ReloadAck():
                     pass
                 case WorkerError(message=msg):
+                    self.current_model_id = None
                     raise RuntimeError(f"Model reload failed: {msg}")
                 case _:
+                    self.current_model_id = None
                     raise RuntimeError(f"Unexpected reload response: "
                                        f"{type(reload_response).__name__}")
 

apps/dreamverse/dreamverse/ltx2_generation.py

@@ -1,9 +1,9 @@
-"""LTX2 model lifecycle and continuation conditioning.
+"""LTX-2 model lifecycle and continuation conditioning.
 
 Runs inside a GPU worker subprocess.  Owns the model, the audio
 encoder, and the per-session continuation state carried across
 segments.  Callers must set ``os.environ["CUDA_VISIBLE_DEVICES"]``
-before constructing ``VideoGenerationWorker`` — all ``fastvideo.*``
+before constructing ``LTX2GenerationBackend`` — all ``fastvideo.*``
 imports are deferred to method bodies so nothing touches CUDA at
 module import time.
 """
@@ -14,9 +14,6 @@ import gc
 import os
 import re
 import time
-from dataclasses import dataclass
-from typing import Any
-
 import numpy as np
 import torch
 
@@ -35,6 +32,7 @@ from dreamverse.config import (
     DREAMVERSE_LORA_STACK,
     _resolve_lora_spec,
 )
+from dreamverse.generation_contracts import StepResult
 
 # Multi-frame decoded continuation defaults from
 # examples/inference/basic/basic_ltx2_distilled_video_continuation.py.
@@ -80,22 +78,6 @@ def _reset_lora_registry(worker) -> dict:
     return {"status": "lora_registry_reset"}
 
 
-@dataclass
-class StepResult:
-    """Output of one generation step.
-
-    ``head_trim_frames`` / ``head_trim_audio_frames`` are derived here
-    so downstream AV streaming never needs to import conditioning
-    constants.
-    """
-    frames: list
-    audio: Any
-    audio_sample_rate: int | None
-    timings: dict
-    head_trim_frames: int
-    head_trim_audio_frames: int
-
-
 class ContinuationState:
     """Per-session video + audio conditioning carried across segments."""
 
@@ -202,7 +184,7 @@ class ContinuationState:
         self.audio_latents = latents.detach().clone().cpu()
 
 
-class VideoGenerationWorker:
+class LTX2GenerationBackend:
     """Single-GPU LTX2 generator with continuation state.
 
     Caller must set ``os.environ["CUDA_VISIBLE_DEVICES"]`` before

apps/dreamverse/dreamverse/minimax_h3_generation.py

@@ -0,0 +1,297 @@
+"""FastH3 model lifecycle and first-frame continuation for DreamVerse."""
+
+from __future__ import annotations
+
+import gc
+import os
+import time
+from typing import TYPE_CHECKING, Any
+
+import numpy as np
+import torch
+
+from dreamverse.config import DREAMVERSE_SP_SIZE
+from dreamverse.generation_contracts import StepResult
+
+if TYPE_CHECKING:
+    from PIL.Image import Image
+
+
+def _required_config_str(model_config: dict, field_name: str) -> str:
+    """Read one required non-empty string from a DreamVerse model profile."""
+    value = model_config.get(field_name)
+    if not isinstance(value, str) or not value.strip():
+        raise ValueError(f"FastH3 model configuration requires `{field_name}`.")
+    return value.strip()
+
+
+class MiniMaxH3GenerationBackend:
+    """Run the VSA data-free FastH3 adapter and retain one continuation frame."""
+
+    def __init__(self, gpu_id: int):
+        self.gpu_id = gpu_id
+        self.generator: Any | None = None
+        self.model_config: dict = {}
+        self.continuation_image: Image | None = None
+
+    def _gpu_mem(self) -> str:
+        allocated_gib = torch.cuda.memory_allocated() / 1024**3
+        reserved_gib = torch.cuda.memory_reserved() / 1024**3
+        return f"alloc={allocated_gib:.2f}GiB, reserved={reserved_gib:.2f}GiB"
+
+    @staticmethod
+    def _configure_environment(attention_backend: str) -> None:
+        """Apply the fixed boot-time switches from the FastH3 reference recipe."""
+        os.environ.update({
+            "FASTVIDEO_ATTENTION_BACKEND": attention_backend,
+            "FASTVIDEO_FA4": "1",
+            "FASTVIDEO_MINIMAX_H3_FUSIONS": "all",
+            "FASTVIDEO_VSA_SM100A": "0",
+        })
+        os.environ.pop("FASTVIDEO_INFERENCE_TORCH_COMPILE", None)
+
+    def initialize(self, model_config: dict | None = None) -> None:
+        """Download the fixed Preview adapter and load the FastH3 generator.
+
+        The model profile owns the base checkpoint, adapter file, attention
+        backend, and generation geometry. The backend translates that profile
+        into FastVideo's typed generator configuration.
+        """
+        if model_config is not None:
+            self.model_config = dict(model_config)
+        if not self.model_config:
+            raise ValueError("FastH3 initialization requires a model configuration.")
+
+        if self.generator is not None:
+            self.generator.shutdown()
+            self.generator = None
+            gc.collect()
+            torch.cuda.empty_cache()
+
+        self.clear_conditioning()
+        model_path = _required_config_str(self.model_config, "model_path")
+        adapter_repo = _required_config_str(self.model_config, "adapter_repo")
+        adapter_filename = _required_config_str(self.model_config, "adapter_filename")
+        attention_backend = _required_config_str(self.model_config, "attention_backend")
+        self._configure_environment(attention_backend)
+
+        from huggingface_hub import hf_hub_download
+
+        from fastvideo import VideoGenerator
+        from fastvideo.api import (
+            CompileConfig,
+            ComponentConfig,
+            EngineConfig,
+            GeneratorConfig,
+            OffloadConfig,
+            ParallelismConfig,
+            PipelineSelection,
+        )
+
+        adapter_path = hf_hub_download(repo_id=adapter_repo, filename=adapter_filename)
+        experimental = {
+            "attention_backend": attention_backend,
+            "inference_torch_compile": attention_backend == "FLASH_ATTN",
+            "vae_parallel_decode": True,
+            "vae_parallel_decode_strategy": "gather",
+        }
+        if attention_backend == "VIDEO_SPARSE_ATTN_H3":
+            experimental.update({
+                "VSA_sparsity": 0.9,
+                "VSA_tile_size": 64,
+            })
+        generator_config = GeneratorConfig(
+            model_path=model_path,
+            pipeline=PipelineSelection(
+                components=ComponentConfig(lora_path=adapter_path, lora_strength=1.0),
+                experimental=experimental,
+            ),
+            engine=EngineConfig(
+                num_gpus=DREAMVERSE_SP_SIZE,
+                parallelism=ParallelismConfig(tp_size=1, sp_size=DREAMVERSE_SP_SIZE),
+                offload=OffloadConfig(
+                    dit=False,
+                    dit_layerwise=False,
+                    text_encoder=True,
+                    image_encoder=True,
+                    vae=True,
+                    pin_cpu_memory=True,
+                ),
+                compile=CompileConfig(enabled=False, vae_enabled=True),
+                use_fsdp_inference=False,
+            ),
+        )
+
+        print(f"[GPU {self.gpu_id}] Loading FastH3 model: {model_path}")
+        print(f"[GPU {self.gpu_id}] FastH3 adapter: {adapter_repo}/{adapter_filename}")
+        print(f"[GPU {self.gpu_id}] Before model load: {self._gpu_mem()}")
+        self.generator = VideoGenerator.from_config(generator_config)
+        print(f"[GPU {self.gpu_id}] FastH3 loaded: {self._gpu_mem()} (warmup pending)")
+
+    def shutdown(self) -> None:
+        """Release the FastVideo generator and cached continuation image."""
+        self.clear_conditioning()
+        if self.generator is not None:
+            self.generator.shutdown()
+            self.generator = None
+
+    def clear_conditioning(self) -> None:
+        """Release the first-frame image retained for the next segment."""
+        if self.continuation_image is not None:
+            self.continuation_image.close()
+            self.continuation_image = None
+
+    @staticmethod
+    def _load_rgb_image(image_path: str) -> Image:
+        """Load an image into an independent RGB buffer with no open file handle."""
+        from PIL import Image
+
+        with Image.open(image_path) as image:
+            return image.convert("RGB").copy()
+
+    def _select_conditioning_image(
+        self,
+        segment_idx: int,
+        image_path: str | None,
+        reset_conditioning: bool,
+    ) -> tuple[Image | None, bool]:
+        """Select the initial upload or retained last frame for one segment."""
+        if reset_conditioning:
+            self.clear_conditioning()
+        if segment_idx > 1 and self.continuation_image is not None:
+            return self.continuation_image.copy(), True
+        if segment_idx > 1 and not reset_conditioning:
+            raise RuntimeError(f"FastH3 segment {segment_idx} requires a retained continuation frame.")
+        if segment_idx == 1 and image_path:
+            return self._load_rgb_image(image_path), False
+        return None, False
+
+    def _build_request(self, prompt: str, conditioning_image: Image | None):
+        """Build the typed FastVideo request owned by the FastH3 profile."""
+        from fastvideo.api import GenerationRequest, InputConfig, OutputConfig, SamplingConfig
+
+        return GenerationRequest(
+            prompt=prompt,
+            negative_prompt="",
+            inputs=InputConfig(pil_image=conditioning_image),
+            sampling=SamplingConfig(
+                height=int(self.model_config["height"]),
+                width=int(self.model_config["width"]),
+                num_frames=int(self.model_config["num_frames"]),
+                fps=24,
+                num_inference_steps=int(self.model_config["num_inference_steps"]),
+                guidance_scale=1.0,
+                batch_cfg=False,
+                seed=int(self.model_config["seed"]),
+            ),
+            output=OutputConfig(save_video=False, return_frames=True),
+        )
+
+    def _save_continuation_frame(self, frames: list) -> None:
+        """Retain the last decoded frame as first-frame conditioning."""
+        from PIL import Image
+
+        self.clear_conditioning()
+        self.continuation_image = Image.fromarray(np.ascontiguousarray(frames[-1])).convert("RGB")
+
+    def generate_step(
+        self,
+        prompt: str,
+        segment_idx: int,
+        image_path: str | None,
+        reset_conditioning: bool,
+    ) -> StepResult:
+        """Generate one synchronized FastH3 segment and retain its last frame.
+
+        Later segments use MiniMax H3's first-frame-to-video path. The first
+        conditioned frame and its matching audio duration are trimmed before
+        streaming so adjacent segments do not duplicate media.
+        """
+        if self.generator is None:
+            raise RuntimeError("FastH3 generator is not initialized.")
+        conditioning_image, uses_continuation = self._select_conditioning_image(
+            segment_idx,
+            image_path,
+            reset_conditioning,
+        )
+        request = self._build_request(prompt, conditioning_image)
+        started = time.perf_counter()
+        try:
+            result = self.generator.generate(request)
+        finally:
+            if conditioning_image is not None:
+                conditioning_image.close()
+        torch.cuda.synchronize()
+        generation_ms = (time.perf_counter() - started) * 1000.0
+
+        if isinstance(result, list):
+            raise RuntimeError("FastH3 returned multiple results for one DreamVerse segment.")
+        frames = result.frames
+        if not isinstance(frames, list) or not frames:
+            raise RuntimeError("FastH3 generation did not return decoded frames.")
+        audio = result.audio
+        audio_sample_rate = result.audio_sample_rate
+        if audio is not None and audio_sample_rate is None:
+            raise RuntimeError("FastH3 returned audio without an audio sample rate.")
+
+        save_started = time.perf_counter()
+        self._save_continuation_frame(frames)
+        save_conditioning_ms = (time.perf_counter() - save_started) * 1000.0
+        timings = {
+            "generation_ms": generation_ms,
+            "generation_time_ms": float(result.generation_time or 0.0) * 1000.0,
+            "save_conditioning_ms": save_conditioning_ms,
+            "e2e_latency_ms": (time.perf_counter() - started) * 1000.0,
+        }
+        trim_frames = 1 if uses_continuation else 0
+        print(f"[GPU {self.gpu_id}] FastH3 segment {segment_idx}: "
+              f"{len(frames)} frames, gen={generation_ms:.0f}ms, "
+              f"save_conditioning={save_conditioning_ms:.0f}ms, "
+              f"e2e={timings['e2e_latency_ms']:.0f}ms")
+        return StepResult(
+            frames=frames,
+            audio=audio,
+            audio_sample_rate=audio_sample_rate,
+            timings=timings,
+            head_trim_frames=trim_frames,
+            head_trim_audio_frames=trim_frames,
+        )
+
+    def warmup(self, prompt: str) -> dict[str, float]:
+        """Compile the FastH3 text and first-frame paths before readiness."""
+        warmup_prompt = (prompt or "").strip()
+        if not warmup_prompt:
+            raise RuntimeError("Startup warmup prompt must be non-empty.")
+        print(f"[GPU {self.gpu_id}] FastH3 startup warmup starting "
+              "(synthetic segments: text-to-video, first-frame-to-video)")
+        started = time.perf_counter()
+        text_result = self.generate_step(
+            warmup_prompt,
+            segment_idx=1,
+            image_path=None,
+            reset_conditioning=True,
+        )
+        first_frame_result = self.generate_step(
+            warmup_prompt,
+            segment_idx=2,
+            image_path=None,
+            reset_conditioning=False,
+        )
+        total_ms = (time.perf_counter() - started) * 1000.0
+        self.clear_conditioning()
+        text_ms = float(text_result.timings.get("e2e_latency_ms", 0.0))
+        first_frame_ms = float(first_frame_result.timings.get("e2e_latency_ms", 0.0))
+        print(f"[GPU {self.gpu_id}] FastH3 startup warmup complete: "
+              f"text_to_video={text_ms:.0f}ms, "
+              f"first_frame_to_video={first_frame_ms:.0f}ms, "
+              f"total={total_ms:.0f}ms")
+        return {
+            "warmup_text_to_video_ms": text_ms,
+            "warmup_first_frame_to_video_ms": first_frame_ms,
+            "warmup_total_ms": total_ms,
+        }
+
+    def apply_lora_stack(self, stack: list[tuple[str, float]]) -> tuple[str | None, str | None]:
+        """Reject runtime LoRA mutation because FastH3 uses one startup adapter."""
+        del stack
+        raise RuntimeError("FastH3 uses its fixed startup adapter and does not support runtime LoRA changes.")

apps/dreamverse/dreamverse/session/controller.py

@@ -30,7 +30,7 @@ from dreamverse.session_init_image import cleanup_session_init_image, persist_se
 from dreamverse.worker_ipc import MediaChunk, MediaComplete, MediaInit
 
 from dreamverse.config import (
-    DEFAULT_MODEL_ID,
+    ACTIVE_MODEL_ID,
     GENERATION_SEGMENT_CAP,
     PROMPT_AUTO_SLEEP_MS,
     PROMPT_AUTO_TIMEOUT_MS,
@@ -264,7 +264,7 @@ class SessionController:
             timeout_task = asyncio.create_task(session_timeout())
 
             # Join the engine on this GPU.
-            await slot.join_user(client_id, model_id=DEFAULT_MODEL_ID)
+            await slot.join_user(client_id, model_id=ACTIVE_MODEL_ID)
 
             # Notify client they're connected to a GPU.
             await ws_send_json({

apps/dreamverse/dreamverse/tests/conftest.py

@@ -3,7 +3,6 @@ from __future__ import annotations
 import sys
 from pathlib import Path
 
-
 TESTS_DIR = Path(__file__).resolve().parent
 DREAMVERSE_PACKAGE_DIR = TESTS_DIR.parent
 DREAMVERSE_APP_DIR = DREAMVERSE_PACKAGE_DIR.parent

apps/dreamverse/dreamverse/tests/test_config.py

@@ -2,14 +2,14 @@ from __future__ import annotations
 
 import importlib.util
 from pathlib import Path
+from types import ModuleType
 
 import pytest
 
-
 SERVER_DIR = Path(__file__).resolve().parents[1]
 
 
-def _load_config_module():
+def _load_config_module() -> ModuleType:
     spec = importlib.util.spec_from_file_location(
         "server_config_test_module",
         SERVER_DIR / "config.py",
@@ -21,7 +21,7 @@ def _load_config_module():
     return module
 
 
-def _set_required_prompt_keys(monkeypatch):
+def _set_required_prompt_keys(monkeypatch: pytest.MonkeyPatch) -> None:
     monkeypatch.setenv("CEREBRAS_API_KEY", "cerebras-key")
     monkeypatch.setenv("GROQ_API_KEY", "groq-key")
 
@@ -53,9 +53,7 @@ def test_config_defaults_to_cerebras_with_parallel_groq_fallback_stage(monkeypat
     module = _load_config_module()
 
     assert module.PROMPT_PROVIDER == "cerebras"
-    assert module.PROMPT_PROVIDER_RUNTIME_STAGES == (
-        ("cerebras", "groq"),
-    )
+    assert module.PROMPT_PROVIDER_RUNTIME_STAGES == (("cerebras", "groq"), )
     assert module.PROMPT_PROVIDER_PRIORITY == (
         "cerebras",
         "groq",
@@ -86,9 +84,7 @@ def test_config_ignores_legacy_groq_primary_override(monkeypatch):
     module = _load_config_module()
 
     assert module.PROMPT_PROVIDER == "cerebras"
-    assert module.PROMPT_PROVIDER_RUNTIME_STAGES == (
-        ("cerebras", "groq"),
-    )
+    assert module.PROMPT_PROVIDER_RUNTIME_STAGES == (("cerebras", "groq"), )
     assert module.PROMPT_PROVIDER_PRIORITY == (
         "cerebras",
         "groq",
@@ -106,24 +102,17 @@ def test_config_uses_local_overlay_paths_when_devtools_enabled(monkeypatch, tmp_
 
     assert module.DEVTOOLS_ENABLED is True
     assert module.FRONTEND_ROOT.as_posix().endswith("apps/dreamverse/web")
-    assert module.PROMPT_ENHANCE_SYSTEM_PROMPT_PATH.endswith(
-        "dreamverse/prompts.local/next_segment_system_prompt.md"
-    )
+    assert module.PROMPT_ENHANCE_SYSTEM_PROMPT_PATH.endswith("dreamverse/prompts.local/next_segment_system_prompt.md")
     assert module.PROMPT_ENHANCE_SYSTEM_PROMPT_FALLBACK_PATH.endswith(
-        "dreamverse/prompts/next_segment_system_prompt.md"
-    )
+        "dreamverse/prompts/next_segment_system_prompt.md")
     assert module.PROMPT_REWRITE_USER_SYSTEM_PROMPT_PATH.endswith(
-        "dreamverse/prompts.local/rewrite_user_system_prompt.md"
-    )
+        "dreamverse/prompts.local/rewrite_user_system_prompt.md")
     assert module.PROMPT_REWRITE_USER_SYSTEM_PROMPT_FALLBACK_PATH.endswith(
-        "dreamverse/prompts/rewrite_user_system_prompt.md"
-    )
+        "dreamverse/prompts/rewrite_user_system_prompt.md")
     assert module.CURATED_PRESETS_FILE_PATH.endswith(
-        "apps/dreamverse/web/prompts.local/selected_ltx2_continuation_story_presets.json"
-    )
+        "apps/dreamverse/web/prompts.local/selected_ltx2_continuation_story_presets.json")
     assert module.CURATED_PRESETS_FALLBACK_FILE_PATH.endswith(
-        "apps/dreamverse/web/prompts/selected_ltx2_continuation_story_presets.json"
-    )
+        "apps/dreamverse/web/prompts/selected_ltx2_continuation_story_presets.json")
     assert module.FRONTEND_STATIC_DIR_CANDIDATES[:2] == (
         str(module.FRONTEND_ROOT / "out"),
         str(module.FRONTEND_ROOT / "dist"),
@@ -162,3 +151,38 @@ def test_config_rejects_invalid_prompt_provider(monkeypatch):
 
     with pytest.raises(RuntimeError, match="Invalid FASTVIDEO_PROMPT_PROVIDER"):
         _load_config_module()
+
+
+def test_config_registers_vsa_datafree_fasth3_profile(monkeypatch):
+    """The FastH3 registry entry owns the complete fixed Preview recipe."""
+    _set_required_prompt_keys(monkeypatch)
+
+    module = _load_config_module()
+
+    assert module.MODEL_REGISTRY["fast-h3"] == {
+        "name": "FastH3",
+        "generation_backend": "minimax_h3",
+        "default_sp_size": 4,
+        "model_path": "MiniMaxAI/MiniMax-H3",
+        "adapter_repo": "FastVideo/FastVideo-FastH3-4-step-Preview-v1-LoRA",
+        "adapter_filename": "vsa-datafree/adapter_model.safetensors",
+        "attention_backend": "VIDEO_SPARSE_ATTN_H3",
+        "height": 768,
+        "width": 1344,
+        "num_frames": 124,
+        "num_inference_steps": 5,
+        "seed": 1000,
+    }
+
+
+def test_config_uses_fasth3_sequence_parallel_default(monkeypatch):
+    """Selecting FastH3 defaults DreamVerse to its four-GPU topology."""
+    _set_required_prompt_keys(monkeypatch)
+    monkeypatch.setenv("DREAMVERSE_MODEL_ID", "fast-h3")
+    monkeypatch.delenv("DREAMVERSE_SP_SIZE", raising=False)
+
+    module = _load_config_module()
+
+    assert module.ACTIVE_MODEL_ID == "fast-h3"
+    assert module.MODEL_CONFIG["generation_backend"] == "minimax_h3"
+    assert module.DREAMVERSE_SP_SIZE == 4

apps/dreamverse/dreamverse/tests/test_entrypoints.py

@@ -9,6 +9,7 @@ from fastapi.testclient import TestClient
 import fastvideo.entrypoints.streaming as streaming_entrypoints
 import pytest
 
+
 def _install_stack03_import_stubs(monkeypatch):
     """Keep entrypoint tests focused while later-stack runtime modules are absent."""
     if not hasattr(streaming_entrypoints, "build_health_router"):
@@ -17,6 +18,7 @@ def _install_stack03_import_stubs(monkeypatch):
     gpu_pool_stub = types.ModuleType("dreamverse.gpu_pool")
 
     class GPUPool:
+
         def __init__(self, _gpu_ids):
             pass
 
@@ -49,6 +51,7 @@ def _install_stack03_import_stubs(monkeypatch):
     controller_stub = types.ModuleType("dreamverse.session.controller")
 
     class SessionController:
+
         def __init__(self, **_kwargs):
             pass
 
@@ -76,13 +79,11 @@ def _run_cli(module, monkeypatch, argv: list[str]) -> list[dict[str, object]]:
     uvicorn_stub = types.ModuleType("uvicorn")
 
     def run(app, host: str, port: int) -> None:
-        calls.append(
-            {
-                "app": app,
-                "host": host,
-                "port": port,
-            }
-        )
+        calls.append({
+            "app": app,
+            "host": host,
+            "port": port,
+        })
 
     uvicorn_stub.run = run
     monkeypatch.setitem(sys.modules, "uvicorn", uvicorn_stub)
@@ -99,13 +100,11 @@ def test_server_cli_defaults_to_local_web_port(monkeypatch):
     server_main = _import_server_main(monkeypatch)
     calls = _run_cli(server_main, monkeypatch, ["dreamverse-server"])
 
-    assert calls == [
-        {
-            "app": server_main.app,
-            "host": "0.0.0.0",
-            "port": 8009,
-        }
-    ]
+    assert calls == [{
+        "app": server_main.app,
+        "host": "0.0.0.0",
+        "port": 8009,
+    }]
 
 
 def test_server_cli_allows_explicit_host_and_port(monkeypatch):
@@ -116,13 +115,11 @@ def test_server_cli_allows_explicit_host_and_port(monkeypatch):
         ["dreamverse-server", "--host", "127.0.0.1", "--port", "8123"],
     )
 
-    assert calls == [
-        {
-            "app": server_main.app,
-            "host": "127.0.0.1",
-            "port": 8123,
-        }
-    ]
+    assert calls == [{
+        "app": server_main.app,
+        "host": "127.0.0.1",
+        "port": 8123,
+    }]
 
 
 def test_server_does_not_expose_backend_source_as_static_assets(monkeypatch):
@@ -142,13 +139,11 @@ def test_mock_server_cli_defaults_to_local_web_port(monkeypatch):
         ["dreamverse-mock-server"],
     )
 
-    assert calls == [
-        {
-            "app": mock_server.app,
-            "host": "0.0.0.0",
-            "port": 8009,
-        }
-    ]
+    assert calls == [{
+        "app": mock_server.app,
+        "host": "0.0.0.0",
+        "port": 8009,
+    }]
 
 
 def test_mock_server_cli_updates_latency(monkeypatch):
@@ -161,13 +156,11 @@ def test_mock_server_cli_updates_latency(monkeypatch):
             ["dreamverse-mock-server", "--latency", "321", "--port", "8111"],
         )
 
-        assert calls == [
-            {
-                "app": mock_server.app,
-                "host": "0.0.0.0",
-                "port": 8111,
-            }
-        ]
+        assert calls == [{
+            "app": mock_server.app,
+            "host": "0.0.0.0",
+            "port": 8111,
+        }]
         assert mock_server.LATENCY_MS == 321
     finally:
         mock_server.LATENCY_MS = old_latency_ms

apps/dreamverse/dreamverse/tests/test_generation_worker.py

@@ -0,0 +1,9 @@
+from dreamverse.generation_worker import _create_generation_backend
+from dreamverse.ltx2_generation import LTX2GenerationBackend
+
+
+def test_create_generation_backend_ltx2_module_import():
+    backend = _create_generation_backend("ltx2", gpu_id=3)
+
+    assert isinstance(backend, LTX2GenerationBackend)
+    assert backend.gpu_id == 3

apps/dreamverse/dreamverse/tests/test_gpu_pool.py

@@ -7,7 +7,6 @@ from types import SimpleNamespace
 
 import pytest
 
-
 import dreamverse.gpu_pool as gpu_pool
 
 
@@ -64,6 +63,14 @@ def test_get_available_gpus_defaults_to_first_visible_device(monkeypatch):
     assert gpu_pool.get_available_gpus() == [3]
 
 
+def test_get_available_gpus_defaults_to_active_model_sequence_parallel_size(monkeypatch):
+    monkeypatch.setenv("CUDA_VISIBLE_DEVICES", "0,1,2,3,4")
+    monkeypatch.delenv("FASTVIDEO_GPU_COUNT", raising=False)
+    monkeypatch.setattr(gpu_pool, "DREAMVERSE_SP_SIZE", 4)
+
+    assert gpu_pool.get_available_gpus() == [0, 1, 2, 3]
+
+
 def test_get_available_gpus_rejects_invalid_gpu_count(monkeypatch):
     monkeypatch.delenv("CUDA_VISIBLE_DEVICES", raising=False)
     monkeypatch.setenv("FASTVIDEO_GPU_COUNT", "zero")
@@ -72,6 +79,23 @@ def test_get_available_gpus_rejects_invalid_gpu_count(monkeypatch):
         gpu_pool.get_available_gpus()
 
 
+def test_join_user_failed_reload_marks_model_uninitialized(monkeypatch):
+    """A failed model reload forces the next join to reload a model."""
+    slot = gpu_pool.GPUSlot(gpu_id=0, cuda_device="0")
+    slot.current_model_id = "fast-ltx2"
+
+    async def fake_send_command(command, timeout):
+        del command, timeout
+        return gpu_pool.WorkerError(user_id="__reload__", message="load failed")
+
+    monkeypatch.setattr(slot, "_send_command", fake_send_command)
+
+    with pytest.raises(RuntimeError, match="Model reload failed"):
+        asyncio.run(slot.join_user("client-id", model_id="fast-h3"))
+
+    assert slot.current_model_id is None
+
+
 def test_send_command_raises_on_worker_death():
     """A worker that consumes a command and exits without replying must
     surface as RuntimeError via sentinel detection, not after the long
@@ -85,9 +109,7 @@ def test_send_command_raises_on_worker_death():
     cmd_q = ctx.Queue()
     resp_q = ctx.Queue()
 
-    proc = ctx.Process(
-        target=_child_consume_and_exit, args=(cmd_q, resp_q)
-    )
+    proc = ctx.Process(target=_child_consume_and_exit, args=(cmd_q, resp_q))
     proc.start()
 
     # Wait for the spawn child to fully boot.  Allow generous time —
@@ -95,9 +117,9 @@ def test_send_command_raises_on_worker_death():
     ready = resp_q.get(timeout=30.0)
     assert ready == "READY"
 
-    async def runner():
+    async def runner() -> None:
         slot = gpu_pool.GPUSlot(gpu_id=0, cuda_device="0")
-        slot.process = proc
+        slot.process = proc  # type: ignore[assignment]
         slot.command_queue = cmd_q
         slot.response_queue = resp_q
 

apps/dreamverse/dreamverse/tests/test_import_contract.py

@@ -7,7 +7,7 @@ ALLOWED_PREFIXES = (
     "fastvideo.entrypoints.video_generator",
     "fastvideo.configs",
 )
-ALLOWED_EXACT = ("fastvideo",)
+ALLOWED_EXACT = ("fastvideo", )
 FORBIDDEN_PREFIXES = (
     "fastvideo.pipelines",
     "fastvideo.models",
@@ -17,11 +17,11 @@ FORBIDDEN_PREFIXES = (
 )
 ALLOWED_INTERNAL_IMPORTS = {
     (
-        "video_generation.py",
+        "ltx2_generation.py",
         "fastvideo.models.audio.ltx2_audio_processing",
     ),
     (
-        "video_generation.py",
+        "ltx2_generation.py",
         "fastvideo.models.loader.component_loader",
     ),
 }
@@ -38,19 +38,13 @@ def test_dreamverse_server_imports_only_public_fastvideo_surfaces() -> None:
         except SyntaxError as task_exc:
             raise AssertionError(f"Failed to parse {path}") from task_exc
         for node in ast.walk(tree):
-            names = (
-                [a.name for a in node.names] if isinstance(node, ast.Import)
-                else [node.module] if isinstance(node, ast.ImportFrom) and node.module
-                else []
-            )
+            names = ([a.name for a in node.names] if isinstance(node, ast.Import) else
+                     [node.module] if isinstance(node, ast.ImportFrom) and node.module else [])
             for name in names:
                 if not name:
                     continue
                 rel_path = str(path.relative_to(root))
-                if (
-                    name.startswith(FORBIDDEN_PREFIXES)
-                    and (rel_path, name) not in ALLOWED_INTERNAL_IMPORTS
-                ):
+                if (name.startswith(FORBIDDEN_PREFIXES) and (rel_path, name) not in ALLOWED_INTERNAL_IMPORTS):
                     bad.append((str(path.relative_to(root)), getattr(node, "lineno", 0), name))
 
     assert bad == [], f"Forbidden internal imports: {bad}"

apps/dreamverse/dreamverse/tests/test_minimax_h3_generation.py

@@ -0,0 +1,247 @@
+from __future__ import annotations
+
+import os
+from types import SimpleNamespace
+from typing import Any
+
+import numpy as np
+import pytest
+
+import dreamverse.generation_worker as generation_worker
+from dreamverse.minimax_h3_generation import MiniMaxH3GenerationBackend
+
+
+FASTH3_MODEL_CONFIG = {
+    "name": "FastH3",
+    "generation_backend": "minimax_h3",
+    "default_sp_size": 4,
+    "model_path": "MiniMaxAI/MiniMax-H3",
+    "adapter_repo": "FastVideo/FastVideo-FastH3-4-step-Preview-v1-LoRA",
+    "adapter_filename": "vsa-datafree/adapter_model.safetensors",
+    "attention_backend": "VIDEO_SPARSE_ATTN_H3",
+    "height": 768,
+    "width": 1344,
+    "num_frames": 124,
+    "num_inference_steps": 5,
+    "seed": 1000,
+}
+
+
+class _RecordingGenerator:
+    """Record typed requests and return small synchronized media fixtures."""
+
+    def __init__(self) -> None:
+        self.requests: list[Any] = []
+        self.conditioning_pixels: list[np.ndarray | None] = []
+
+    def generate(self, request):
+        """Capture the request and return two tiny video frames with audio."""
+        self.requests.append(request)
+        conditioning_image = request.inputs.pil_image
+        self.conditioning_pixels.append(
+            None if conditioning_image is None else np.asarray(conditioning_image).copy())
+        frames = [
+            np.full((2, 3, 3), 10, dtype=np.uint8),
+            np.full((2, 3, 3), 20, dtype=np.uint8),
+        ]
+        return SimpleNamespace(
+            frames=frames,
+            audio=np.zeros((2, 16), dtype=np.float32),
+            audio_sample_rate=44100,
+            generation_time=0.25,
+        )
+
+
+def test_initialize_builds_vsa_datafree_fasth3_generator(monkeypatch):
+    """Initialization translates the DreamVerse profile into typed FastVideo config."""
+    from fastvideo import VideoGenerator
+
+    captured = {}
+    fake_generator = SimpleNamespace(shutdown=lambda: None)
+
+    def fake_from_config(config):
+        captured["config"] = config
+        return fake_generator
+
+    def fake_download(**kwargs):
+        captured["download"] = kwargs
+        return f"/models/{kwargs['filename']}"
+
+    monkeypatch.setattr("huggingface_hub.hf_hub_download", fake_download)
+    monkeypatch.setattr(VideoGenerator, "from_config", fake_from_config)
+    monkeypatch.setattr("dreamverse.minimax_h3_generation.DREAMVERSE_SP_SIZE", 4)
+    monkeypatch.setenv("FASTVIDEO_ATTENTION_BACKEND", "test-attention")
+    monkeypatch.setenv("FASTVIDEO_FA4", "0")
+    monkeypatch.setenv("FASTVIDEO_MINIMAX_H3_FUSIONS", "0")
+    monkeypatch.setenv("FASTVIDEO_VSA_SM100A", "1")
+    monkeypatch.setenv("FASTVIDEO_INFERENCE_TORCH_COMPILE", "1")
+
+    backend = MiniMaxH3GenerationBackend(gpu_id=0)
+    monkeypatch.setattr(backend, "_gpu_mem", lambda: "alloc=0.00GiB, reserved=0.00GiB")
+    backend.initialize(FASTH3_MODEL_CONFIG)
+
+    config = captured["config"]
+    assert captured["download"] == {
+        "repo_id": "FastVideo/FastVideo-FastH3-4-step-Preview-v1-LoRA",
+        "filename": "vsa-datafree/adapter_model.safetensors",
+    }
+    assert config.model_path == "MiniMaxAI/MiniMax-H3"
+    assert config.pipeline.components.lora_path.endswith("vsa-datafree/adapter_model.safetensors")
+    assert config.pipeline.components.lora_strength == 1.0
+    assert config.pipeline.experimental == {
+        "attention_backend": "VIDEO_SPARSE_ATTN_H3",
+        "inference_torch_compile": False,
+        "vae_parallel_decode": True,
+        "vae_parallel_decode_strategy": "gather",
+        "VSA_sparsity": 0.9,
+        "VSA_tile_size": 64,
+    }
+    assert config.engine.num_gpus == 4
+    assert config.engine.parallelism.tp_size == 1
+    assert config.engine.parallelism.sp_size == 4
+    assert config.engine.offload.dit is False
+    assert config.engine.offload.dit_layerwise is False
+    assert config.engine.offload.text_encoder is True
+    assert config.engine.offload.vae is True
+    assert config.engine.compile.vae_enabled is True
+    assert config.engine.use_fsdp_inference is False
+    assert os.environ["FASTVIDEO_ATTENTION_BACKEND"] == "VIDEO_SPARSE_ATTN_H3"
+    assert os.environ["FASTVIDEO_FA4"] == "1"
+    assert os.environ["FASTVIDEO_MINIMAX_H3_FUSIONS"] == "all"
+    assert os.environ["FASTVIDEO_VSA_SM100A"] == "0"
+    assert "FASTVIDEO_INFERENCE_TORCH_COMPILE" not in os.environ
+
+
+def test_initialize_selects_declared_generation_backend(monkeypatch):
+    """The GPU worker constructs the backend that the active model profile declares."""
+    from unittest.mock import Mock
+
+    selected_backend = Mock()
+    monkeypatch.setattr(
+        generation_worker,
+        "_create_generation_backend",
+        lambda backend_name, gpu_id: selected_backend,
+    )
+    worker = generation_worker.VideoGenerationWorker(gpu_id=3)
+
+    worker.initialize(FASTH3_MODEL_CONFIG)
+
+    assert worker.backend_name == "minimax_h3"
+    assert worker.backend is selected_backend
+    selected_backend.initialize.assert_called_once_with(FASTH3_MODEL_CONFIG)
+
+
+def test_initialize_failure_clears_backend_ownership(monkeypatch):
+    """A failed family change leaves the GPU worker explicitly uninitialized."""
+    ltx_backend = SimpleNamespace(initialize=lambda config: None, shutdown=lambda: None)
+
+    def fail_initialize(config):
+        del config
+        raise RuntimeError("load failed")
+
+    fasth3_backend = SimpleNamespace(
+        initialize=fail_initialize,
+        shutdown=lambda: None,
+    )
+    backends = {
+        "ltx2": ltx_backend,
+        "minimax_h3": fasth3_backend,
+    }
+    monkeypatch.setattr(
+        generation_worker,
+        "_create_generation_backend",
+        lambda backend_name, gpu_id: backends[backend_name],
+    )
+    worker = generation_worker.VideoGenerationWorker(gpu_id=3)
+    worker.initialize({"generation_backend": "ltx2"})
+
+    with pytest.raises(RuntimeError, match="load failed"):
+        worker.initialize(FASTH3_MODEL_CONFIG)
+
+    assert worker.backend is None
+    assert worker.backend_name is None
+    assert worker.model_config == {"generation_backend": "ltx2"}
+
+
+def test_generate_step_uses_last_frame_for_continuation(monkeypatch):
+    """A later segment receives the prior segment's last decoded frame."""
+    backend = MiniMaxH3GenerationBackend(gpu_id=0)
+    backend.model_config = dict(FASTH3_MODEL_CONFIG)
+    backend.generator = _RecordingGenerator()
+    monkeypatch.setattr("dreamverse.minimax_h3_generation.torch.cuda.synchronize", lambda: None)
+
+    first_result = backend.generate_step(
+        "first prompt",
+        segment_idx=1,
+        image_path=None,
+        reset_conditioning=True,
+    )
+    second_result = backend.generate_step(
+        "second prompt",
+        segment_idx=2,
+        image_path=None,
+        reset_conditioning=False,
+    )
+
+    first_request = backend.generator.requests[0]
+    assert first_request.inputs.pil_image is None
+    assert first_request.negative_prompt == ""
+    assert first_request.sampling.height == 768
+    assert first_request.sampling.width == 1344
+    assert first_request.sampling.num_frames == 124
+    assert first_request.sampling.num_inference_steps == 5
+    assert first_request.sampling.fps == 24
+    assert first_request.sampling.guidance_scale == 1.0
+    assert first_request.sampling.batch_cfg is False
+    assert first_request.sampling.seed == 1000
+    assert first_request.output.save_video is False
+    assert first_request.output.return_frames is True
+    assert backend.generator.conditioning_pixels[1].tolist() == np.full((2, 3, 3), 20).tolist()
+    assert first_result.head_trim_frames == 0
+    assert first_result.head_trim_audio_frames == 0
+    assert second_result.head_trim_frames == 1
+    assert second_result.head_trim_audio_frames == 1
+    assert second_result.audio_sample_rate == 44100
+
+
+def test_generate_step_reset_uses_text_to_video_path(monkeypatch):
+    """Resetting continuation produces an unconditioned text-to-video request."""
+    backend = MiniMaxH3GenerationBackend(gpu_id=0)
+    backend.model_config = dict(FASTH3_MODEL_CONFIG)
+    backend.generator = _RecordingGenerator()
+    monkeypatch.setattr("dreamverse.minimax_h3_generation.torch.cuda.synchronize", lambda: None)
+
+    backend.generate_step("first prompt", 1, None, True)
+    reset_result = backend.generate_step("reset prompt", 2, None, True)
+
+    assert backend.generator.requests[-1].inputs.pil_image is None
+    assert reset_result.head_trim_frames == 0
+    assert reset_result.head_trim_audio_frames == 0
+
+
+def test_generate_step_missing_continuation_frame(monkeypatch):
+    """A later segment fails when no reset or retained frame defines its input."""
+    backend = MiniMaxH3GenerationBackend(gpu_id=0)
+    backend.model_config = dict(FASTH3_MODEL_CONFIG)
+    backend.generator = _RecordingGenerator()
+
+    with pytest.raises(RuntimeError, match="requires a retained continuation frame"):
+        backend.generate_step("later prompt", 2, None, False)
+
+    assert backend.generator.requests == []
+
+
+def test_warmup_exercises_text_and_first_frame_paths(monkeypatch):
+    """Warmup covers both request shapes used by a DreamVerse session."""
+    backend = MiniMaxH3GenerationBackend(gpu_id=0)
+    backend.model_config = dict(FASTH3_MODEL_CONFIG)
+    backend.generator = _RecordingGenerator()
+    monkeypatch.setattr("dreamverse.minimax_h3_generation.torch.cuda.synchronize", lambda: None)
+
+    timings = backend.warmup("warmup prompt")
+
+    assert backend.generator.conditioning_pixels[0] is None
+    assert backend.generator.conditioning_pixels[1] is not None
+    assert backend.continuation_image is None
+    assert "warmup_text_to_video_ms" in timings
+    assert "warmup_first_frame_to_video_ms" in timings

apps/dreamverse/dreamverse/tests/test_mock_server.py

@@ -6,7 +6,6 @@ import os
 
 from fastapi import WebSocketDisconnect
 
-
 os.environ.setdefault("CEREBRAS_API_KEY", "dummy")
 os.environ.setdefault("GROQ_API_KEY", "dummy")
 
@@ -14,6 +13,7 @@ import dreamverse.mock_server as mock_server
 
 
 class _FakeWebSocket:
+
     def __init__(self, messages: list[tuple[float, dict[str, object]]]):
         self._messages = messages
         self._index = 0
@@ -49,34 +49,34 @@ def test_mock_server_matches_current_single5s_protocol():
         mock_server.MOCK_SEGMENT_BYTES = b"mock-fmp4-bytes"
         mock_server.LATENCY_MS = 1
 
-        ws = _FakeWebSocket(
-            [
-                (
-                    0.0,
-                    {
-                        "type": "session_init_v2",
-                        "preset_id": "simple_prompt_1",
-                        "curated_prompts": ["selected prompt"],
-                        "single_clip_mode": True,
-                        "enhancement_enabled": False,
-                        "auto_extension_enabled": False,
-                        "loop_generation_enabled": False,
-                    },
-                ),
-                (
-                    0.01,
-                    {
-                        "type": "simple_generate",
-                        "preset_id": "simple_custom_prompt",
-                        "prompt_id": "simple_custom_prompt",
-                        "prompt": "custom prompt",
-                        "enhancement_enabled": True,
-                        "initial_image": None,
-                    },
-                ),
-                (0.20, {"type": "leave"}),
-            ]
-        )
+        ws = _FakeWebSocket([
+            (
+                0.0,
+                {
+                    "type": "session_init_v2",
+                    "preset_id": "simple_prompt_1",
+                    "curated_prompts": ["selected prompt"],
+                    "single_clip_mode": True,
+                    "enhancement_enabled": False,
+                    "auto_extension_enabled": False,
+                    "loop_generation_enabled": False,
+                },
+            ),
+            (
+                0.01,
+                {
+                    "type": "simple_generate",
+                    "preset_id": "simple_custom_prompt",
+                    "prompt_id": "simple_custom_prompt",
+                    "prompt": "custom prompt",
+                    "enhancement_enabled": True,
+                    "initial_image": None,
+                },
+            ),
+            (0.20, {
+                "type": "leave"
+            }),
+        ])
 
         asyncio.run(mock_server.websocket_endpoint(ws))
 
@@ -92,24 +92,14 @@ def test_mock_server_matches_current_single5s_protocol():
         assert message_types.count("ltx2_stream_complete") == 2
         assert "prompt_sources_blocked" not in message_types
 
-        segment_start_events = [
-            payload
-            for payload in ws.sent_json
-            if payload["type"] == "ltx2_segment_start"
-        ]
-        gpu_assigned_event = next(
-            payload for payload in ws.sent_json if payload["type"] == "gpu_assigned"
-        )
+        segment_start_events = [payload for payload in ws.sent_json if payload["type"] == "ltx2_segment_start"]
+        gpu_assigned_event = next(payload for payload in ws.sent_json if payload["type"] == "gpu_assigned")
         assert gpu_assigned_event["session_timeout"] == mock_server.SESSION_TIMEOUT_SECONDS
         assert [payload["segment_idx"] for payload in segment_start_events] == [1, 1]
         assert segment_start_events[0]["prompt"] == "selected prompt"
         assert segment_start_events[1]["prompt"] == "custom prompt"
 
-        step_complete_events = [
-            payload
-            for payload in ws.sent_json
-            if payload["type"] == "step_complete"
-        ]
+        step_complete_events = [payload for payload in ws.sent_json if payload["type"] == "step_complete"]
         assert len(step_complete_events) == 2
         assert step_complete_events[0]["latency_ms"] == {
             "total": 121.0,
@@ -134,29 +124,29 @@ def test_mock_server_regular_cap_waits_for_rewrite_rollout():
         mock_server.LATENCY_MS = 1
         mock_server.GENERATION_SEGMENT_CAP = 1
 
-        ws = _FakeWebSocket(
-            [
-                (
-                    0.0,
-                    {
-                        "type": "session_init_v2",
-                        "preset_id": "test_preset",
-                        "curated_prompts": ["segment one"],
-                        "enhancement_enabled": True,
-                        "auto_extension_enabled": False,
-                        "loop_generation_enabled": False,
-                    },
-                ),
-                (
-                    0.02,
-                    {
-                        "type": "rewrite_seed_prompts",
-                        "rewrite_instruction": "start a new rollout",
-                    },
-                ),
-                (0.20, {"type": "leave"}),
-            ]
-        )
+        ws = _FakeWebSocket([
+            (
+                0.0,
+                {
+                    "type": "session_init_v2",
+                    "preset_id": "test_preset",
+                    "curated_prompts": ["segment one"],
+                    "enhancement_enabled": True,
+                    "auto_extension_enabled": False,
+                    "loop_generation_enabled": False,
+                },
+            ),
+            (
+                0.02,
+                {
+                    "type": "rewrite_seed_prompts",
+                    "rewrite_instruction": "start a new rollout",
+                },
+            ),
+            (0.20, {
+                "type": "leave"
+            }),
+        ])
 
         asyncio.run(mock_server.websocket_endpoint(ws))
 
@@ -166,11 +156,7 @@ def test_mock_server_regular_cap_waits_for_rewrite_rollout():
         assert "generation_cap_reached" not in message_types
         assert "prompt_sources_blocked" not in message_types
 
-        segment_start_events = [
-            payload
-            for payload in ws.sent_json
-            if payload["type"] == "ltx2_segment_start"
-        ]
+        segment_start_events = [payload for payload in ws.sent_json if payload["type"] == "ltx2_segment_start"]
         assert [payload["segment_idx"] for payload in segment_start_events] == [1, 1]
         assert segment_start_events[0]["prompt"] == "segment one"
         assert segment_start_events[1]["prompt"] == "segment one [start a new rollout]"
@@ -187,54 +173,40 @@ def test_mock_server_rewrite_during_active_segment_restarts_from_first_rewritten
         mock_server.MOCK_SEGMENT_BYTES = b"mock-fmp4-bytes"
         mock_server.LATENCY_MS = 100
 
-        ws = _FakeWebSocket(
-            [
-                (
-                    0.0,
-                    {
-                        "type": "session_init_v2",
-                        "preset_id": "test_preset",
-                        "curated_prompts": ["segment one", "segment two"],
-                        "enhancement_enabled": True,
-                        "auto_extension_enabled": False,
-                        "loop_generation_enabled": False,
-                    },
-                ),
-                (
-                    0.02,
-                    {
-                        "type": "rewrite_seed_prompts",
-                        "rewrite_instruction": "restart from rewrite",
-                    },
-                ),
-                (0.40, {"type": "leave"}),
-            ]
-        )
+        ws = _FakeWebSocket([
+            (
+                0.0,
+                {
+                    "type": "session_init_v2",
+                    "preset_id": "test_preset",
+                    "curated_prompts": ["segment one", "segment two"],
+                    "enhancement_enabled": True,
+                    "auto_extension_enabled": False,
+                    "loop_generation_enabled": False,
+                },
+            ),
+            (
+                0.02,
+                {
+                    "type": "rewrite_seed_prompts",
+                    "rewrite_instruction": "restart from rewrite",
+                },
+            ),
+            (0.40, {
+                "type": "leave"
+            }),
+        ])
 
         asyncio.run(mock_server.websocket_endpoint(ws))
 
-        segment_start_events = [
-            payload
-            for payload in ws.sent_json
-            if payload["type"] == "ltx2_segment_start"
-        ]
+        segment_start_events = [payload for payload in ws.sent_json if payload["type"] == "ltx2_segment_start"]
         assert [payload["prompt"] for payload in segment_start_events[:2]] == [
             "segment one",
             "segment one [restart from rewrite]",
         ]
-        assert all(
-            payload["prompt"] != "segment two"
-            for payload in segment_start_events[1:]
-        )
-        reset_events = [
-            payload
-            for payload in ws.sent_json
-            if payload.get("type") == "seed_prompts_reset_applied"
-        ]
-        assert any(
-            payload.get("reason") == "rewrite_during_generation"
-            for payload in reset_events
-        )
+        assert all(payload["prompt"] != "segment two" for payload in segment_start_events[1:])
+        reset_events = [payload for payload in ws.sent_json if payload.get("type") == "seed_prompts_reset_applied"]
+        assert any(payload.get("reason") == "rewrite_during_generation" for payload in reset_events)
     finally:
         mock_server.MOCK_SEGMENT_BYTES = old_segment_bytes
         mock_server.LATENCY_MS = old_latency_ms
@@ -247,24 +219,24 @@ def test_mock_server_supports_initial_custom_rollout_prompt():
         mock_server.MOCK_SEGMENT_BYTES = b"mock-fmp4-bytes"
         mock_server.LATENCY_MS = 1
 
-        ws = _FakeWebSocket(
-            [
-                (
-                    0.0,
-                    {
-                        "type": "session_init_v2",
-                        "preset_id": "custom_editable",
-                        "preset_label": "Custom rollout",
-                        "curated_prompts": [],
-                        "initial_rollout_prompt": "A moonbase corridor thriller with flooding",
-                        "enhancement_enabled": True,
-                        "auto_extension_enabled": False,
-                        "loop_generation_enabled": False,
-                    },
-                ),
-                (0.20, {"type": "leave"}),
-            ]
-        )
+        ws = _FakeWebSocket([
+            (
+                0.0,
+                {
+                    "type": "session_init_v2",
+                    "preset_id": "custom_editable",
+                    "preset_label": "Custom rollout",
+                    "curated_prompts": [],
+                    "initial_rollout_prompt": "A moonbase corridor thriller with flooding",
+                    "enhancement_enabled": True,
+                    "auto_extension_enabled": False,
+                    "loop_generation_enabled": False,
+                },
+            ),
+            (0.20, {
+                "type": "leave"
+            }),
+        ])
 
         asyncio.run(mock_server.websocket_endpoint(ws))
 
@@ -276,15 +248,9 @@ def test_mock_server_supports_initial_custom_rollout_prompt():
         assert "ltx2_stream_start" in message_types
         assert "prompt_sources_blocked" not in message_types
 
-        segment_start_events = [
-            payload
-            for payload in ws.sent_json
-            if payload["type"] == "ltx2_segment_start"
-        ]
+        segment_start_events = [payload for payload in ws.sent_json if payload["type"] == "ltx2_segment_start"]
         assert segment_start_events
-        assert segment_start_events[0]["prompt"] == (
-            "A moonbase corridor thriller with flooding [segment 1]"
-        )
+        assert segment_start_events[0]["prompt"] == ("A moonbase corridor thriller with flooding [segment 1]")
     finally:
         mock_server.MOCK_SEGMENT_BYTES = old_segment_bytes
         mock_server.LATENCY_MS = old_latency_ms
@@ -297,35 +263,37 @@ def test_mock_server_can_start_new_project_without_reconnecting():
         mock_server.MOCK_SEGMENT_BYTES = b"mock-fmp4-bytes"
         mock_server.LATENCY_MS = 40
 
-        ws = _FakeWebSocket(
-            [
-                (
-                    0.0,
-                    {
-                        "type": "session_init_v2",
-                        "preset_id": "test_preset",
-                        "curated_prompts": ["segment one"],
-                        "enhancement_enabled": True,
-                        "auto_extension_enabled": False,
-                        "loop_generation_enabled": False,
-                    },
-                ),
-                (0.02, {"type": "end_project_keep_session"}),
-                (
-                    0.20,
-                    {
-                        "type": "project_init_v1",
-                        "preset_id": "test_preset_2",
-                        "preset_label": "Test Preset 2",
-                        "curated_prompts": ["segment two"],
-                        "enhancement_enabled": True,
-                        "auto_extension_enabled": False,
-                        "loop_generation_enabled": False,
-                    },
-                ),
-                (0.40, {"type": "leave"}),
-            ]
-        )
+        ws = _FakeWebSocket([
+            (
+                0.0,
+                {
+                    "type": "session_init_v2",
+                    "preset_id": "test_preset",
+                    "curated_prompts": ["segment one"],
+                    "enhancement_enabled": True,
+                    "auto_extension_enabled": False,
+                    "loop_generation_enabled": False,
+                },
+            ),
+            (0.02, {
+                "type": "end_project_keep_session"
+            }),
+            (
+                0.20,
+                {
+                    "type": "project_init_v1",
+                    "preset_id": "test_preset_2",
+                    "preset_label": "Test Preset 2",
+                    "curated_prompts": ["segment two"],
+                    "enhancement_enabled": True,
+                    "auto_extension_enabled": False,
+                    "loop_generation_enabled": False,
+                },
+            ),
+            (0.40, {
+                "type": "leave"
+            }),
+        ])
 
         asyncio.run(mock_server.websocket_endpoint(ws))
 
@@ -336,16 +304,11 @@ def test_mock_server_can_start_new_project_without_reconnecting():
 
         project_idle_index = message_types.index("project_idle")
         stream_start_indexes = [
-            index for index, message_type in enumerate(message_types)
-            if message_type == "ltx2_stream_start"
+            index for index, message_type in enumerate(message_types) if message_type == "ltx2_stream_start"
         ]
         assert stream_start_indexes[0] < project_idle_index < stream_start_indexes[1]
 
-        segment_start_events = [
-            payload
-            for payload in ws.sent_json
-            if payload["type"] == "ltx2_segment_start"
-        ]
+        segment_start_events = [payload for payload in ws.sent_json if payload["type"] == "ltx2_segment_start"]
         assert [payload["prompt"] for payload in segment_start_events[:2]] == [
             "segment one",
             "segment two",

apps/dreamverse/dreamverse/tests/test_prompt_enhancer.py

@@ -6,7 +6,6 @@ import os
 import re
 import time
 
-
 os.environ.setdefault("CEREBRAS_API_KEY", "dummy")
 os.environ.setdefault("GROQ_API_KEY", "dummy")
 
@@ -22,6 +21,7 @@ from dreamverse.prompt_enhancer import (
 
 
 class _FakeResponse:
+
     def __init__(self, payload: dict):
         self._payload = payload
 
@@ -30,6 +30,7 @@ class _FakeResponse:
 
 
 class _FakeSyncCompletions:
+
     def __init__(self, payload: dict):
         self._payload = payload
 
@@ -38,6 +39,7 @@ class _FakeSyncCompletions:
 
 
 class _FakeSyncClient:
+
     def __init__(self, payload: dict):
         self.chat = type(
             "_FakeChat",
@@ -47,6 +49,7 @@ class _FakeSyncClient:
 
 
 class _DelayedSyncCompletions:
+
     def __init__(self, payload: dict, delay_s: float = 0.0, exc: Exception | None = None):
         self._payload = payload
         self._delay_s = delay_s
@@ -61,29 +64,26 @@ class _DelayedSyncCompletions:
 
 
 class _DelayedSyncClient:
+
     def __init__(self, payload: dict, delay_s: float = 0.0, exc: Exception | None = None):
         self.chat = type(
             "_FakeChat",
             (),
-            {
-                "completions": _DelayedSyncCompletions(
-                    payload,
-                    delay_s=delay_s,
-                    exc=exc,
-                )
-            },
+            {"completions": _DelayedSyncCompletions(
+                payload,
+                delay_s=delay_s,
+                exc=exc,
+            )},
         )()
 
 
 def _chat_payload_with_content(content: str) -> dict:
     return {
-        "choices": [
-            {
-                "message": {
-                    "content": content,
-                }
+        "choices": [{
+            "message": {
+                "content": content,
             }
-        ]
+        }]
     }
 
 
@@ -172,6 +172,7 @@ def _build_staged_enhancer(
 
 
 class _FakeOpenAIClient:
+
     def __init__(self, **kwargs):
         self.kwargs = kwargs
         self.chat = type(
@@ -182,6 +183,7 @@ class _FakeOpenAIClient:
 
 
 class _FakeCerebrasClient:
+
     def __init__(self, **kwargs):
         self.kwargs = kwargs
         self.chat = type(
@@ -192,16 +194,12 @@ class _FakeCerebrasClient:
 
 
 def test_parse_json_response_accepts_fenced_json_with_prose():
-    parsed = _parse_json_response(
-        "Here is the rewrite:\n```json\n{\"segment_prompts\":[\"A\",\"B\"]}\n```\nThanks."
-    )
+    parsed = _parse_json_response("Here is the rewrite:\n```json\n{\"segment_prompts\":[\"A\",\"B\"]}\n```\nThanks.")
     assert parsed == {"segment_prompts": ["A", "B"]}
 
 
 def test_parse_json_response_extracts_first_embedded_object():
-    parsed = _parse_json_response(
-        "Model output:\n{\"segment_prompts\":[\"A\",\"B\"]}\n(complete)"
-    )
+    parsed = _parse_json_response("Model output:\n{\"segment_prompts\":[\"A\",\"B\"]}\n(complete)")
     assert parsed == {"segment_prompts": ["A", "B"]}
 
 
@@ -268,16 +266,12 @@ def test_build_client_supports_groq_provider(monkeypatch):
 
 def test_rewrite_prompt_sequence_accepts_segment_prompts_output():
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     result = asyncio.run(
         enhancer.rewrite_prompt_sequence(
             ["prompt one", "prompt two"],
             rewrite_instruction="make it cinematic",
-        )
-    )
+        ))
     assert result.fallback_used is False
     assert result.error is None
     assert result.rollout_id == "preset_a"
@@ -286,15 +280,12 @@ def test_rewrite_prompt_sequence_accepts_segment_prompts_output():
 
 
 def test_rewrite_prompt_sequence_accepts_legacy_rewritten_prompts_output():
-    enhancer = _build_test_enhancer(
-        _chat_payload_with_content('{"rewritten_prompts":["A","B"]}')
-    )
+    enhancer = _build_test_enhancer(_chat_payload_with_content('{"rewritten_prompts":["A","B"]}'))
     result = asyncio.run(
         enhancer.rewrite_prompt_sequence(
             ["prompt one", "prompt two"],
             rewrite_instruction="make it cinematic",
-        )
-    )
+        ))
     assert result.fallback_used is False
     assert result.error is None
     assert result.rollout_id == "current_rollout"
@@ -303,19 +294,14 @@ def test_rewrite_prompt_sequence_accepts_legacy_rewritten_prompts_output():
 
 
 def test_rewrite_prompt_sequence_accepts_segment_dicts_without_top_level_rollout_metadata():
-    enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"segments":[{"prompt":"A"},{"text":"B"}]}'
-        )
-    )
+    enhancer = _build_test_enhancer(_chat_payload_with_content('{"segments":[{"prompt":"A"},{"text":"B"}]}'))
     result = asyncio.run(
         enhancer.rewrite_prompt_sequence(
             ["prompt one", "prompt two"],
             preset_id="preset_a",
             preset_label="Preset A",
             rewrite_instruction="make it cinematic",
-        )
-    )
+        ))
     assert result.fallback_used is False
     assert result.error is None
     assert result.rollout_id == "preset_a"
@@ -329,14 +315,12 @@ def test_rewrite_prompt_sequence_accepts_numbered_prose_output():
             "The user is asking for a cinematic rewrite.\n\n"
             "1. A dog bounds across the moon's dusty surface, kicking up silver regolith as it chases a rabbit beneath the black sky.\n"
             "2. The rabbit darts around a crater rim while the dog lunges after it, Earth glowing blue in the distance.\n"
-        )
-    )
+        ))
     result = asyncio.run(
         enhancer.rewrite_prompt_sequence(
             ["prompt one", "prompt two"],
             rewrite_instruction="make it cinematic",
-        )
-    )
+        ))
     assert result.fallback_used is False
     assert result.error is None
     assert result.rollout_id == "current_rollout"
@@ -347,29 +331,27 @@ def test_rewrite_prompt_sequence_accepts_numbered_prose_output():
     ]
 
 
-def test_enhance_prompt_prefers_cerebras_before_groq_fallback():
+def test_enhance_prompt_uses_groq_when_it_returns_first():
     enhancer = _build_staged_enhancer(
         cerebras_payload=_chat_payload_with_content('{"prompt":"Cerebras prompt"}'),
         groq_payload=_chat_payload_with_content('{"prompt":"Groq prompt"}'),
-        cerebras_delay_s=0.01,
+        cerebras_delay_s=0.08,
         groq_delay_s=0.01,
     )
 
-    result = asyncio.run(
-        enhancer.enhance_prompt(
-            "A rainy alley at night",
-            mode="single_clip",
-        )
-    )
+    result = asyncio.run(enhancer.enhance_prompt(
+        "A rainy alley at night",
+        mode="single_clip",
+    ))
 
     assert result.fallback_used is False
     assert result.error is None
-    assert result.provider == "cerebras"
+    assert result.provider == "groq"
     assert result.model == "gpt-test"
-    assert result.prompt == "Cerebras prompt"
+    assert result.prompt == "Groq prompt"
     assert enhancer.get_provider_success_counts() == {
-        "cerebras": 1,
-        "groq": 0,
+        "cerebras": 0,
+        "groq": 1,
     }
 
 
@@ -381,12 +363,10 @@ def test_enhance_prompt_uses_groq_when_cerebras_fails():
         groq_delay_s=0.01,
     )
 
-    result = asyncio.run(
-        enhancer.enhance_prompt(
-            "A rainy alley at night",
-            mode="single_clip",
-        )
-    )
+    result = asyncio.run(enhancer.enhance_prompt(
+        "A rainy alley at night",
+        mode="single_clip",
+    ))
 
     assert result.fallback_used is False
     assert result.error is None
@@ -408,13 +388,11 @@ def test_enhance_prompt_can_use_groq_when_cerebras_times_out():
     enhancer.http_timeout_ms = 50
     enhancer.default_timeout_ms = 50
 
-    result = asyncio.run(
-        enhancer.enhance_prompt(
-            "A rainy alley at night",
-            mode="single_clip",
-            timeout_ms=50,
-        )
-    )
+    result = asyncio.run(enhancer.enhance_prompt(
+        "A rainy alley at night",
+        mode="single_clip",
+        timeout_ms=50,
+    ))
 
     assert result.fallback_used is False
     assert result.error is None
@@ -434,12 +412,10 @@ def test_enhance_prompt_can_use_cerebras_when_it_returns_first():
         groq_delay_s=0.08,
     )
 
-    result = asyncio.run(
-        enhancer.enhance_prompt(
-            "A rainy alley at night",
-            mode="single_clip",
-        )
-    )
+    result = asyncio.run(enhancer.enhance_prompt(
+        "A rainy alley at night",
+        mode="single_clip",
+    ))
 
     assert result.fallback_used is False
     assert result.error is None
@@ -453,15 +429,12 @@ def test_enhance_prompt_can_use_cerebras_when_it_returns_first():
 
 
 def test_rewrite_prompt_sequence_keeps_raw_output_on_parse_error():
-    enhancer = _build_test_enhancer(
-        _chat_payload_with_content("I cannot comply with JSON right now.")
-    )
+    enhancer = _build_test_enhancer(_chat_payload_with_content("I cannot comply with JSON right now."))
     result = asyncio.run(
         enhancer.rewrite_prompt_sequence(
             ["prompt one", "prompt two"],
             rewrite_instruction="make it cinematic",
-        )
-    )
+        ))
     assert result.fallback_used is True
     assert "No JSON object found in assistant response." in (result.error or "")
     assert result.raw_response_text == "I cannot comply with JSON right now."
@@ -473,9 +446,7 @@ def test_rewrite_prompt_sequence_keeps_raw_output_on_parse_error():
 def test_rewrite_prompt_sequence_uses_current_rollout_payload_shape():
     enhancer = _build_test_enhancer(
         _chat_payload_with_content(
-            '{"id":"rewritten_rollout","label":"Rewritten Rollout","segment_prompts":["A","B"]}'
-        )
-    )
+            '{"id":"rewritten_rollout","label":"Rewritten Rollout","segment_prompts":["A","B"]}'))
     captured = {
         "body": None,
         "timeout_seconds": None,
@@ -486,8 +457,7 @@ def test_rewrite_prompt_sequence_uses_current_rollout_payload_shape():
         captured["timeout_seconds"] = timeout_seconds
         return (
             _chat_payload_with_content(
-                '{"id":"rewritten_rollout","label":"Rewritten Rollout","segment_prompts":["A","B"]}'
-            ),
+                '{"id":"rewritten_rollout","label":"Rewritten Rollout","segment_prompts":["A","B"]}'),
             '{"id":"rewritten_rollout","label":"Rewritten Rollout","segment_prompts":["A","B"]}',
         )
 
@@ -502,8 +472,7 @@ def test_rewrite_prompt_sequence_uses_current_rollout_payload_shape():
             rewrite_model="gpt-test",
             rewrite_temperature=0.2,
             timeout_ms=800,
-        )
-    )
+        ))
 
     assert result.fallback_used is False
     assert captured["body"]["messages"][0] == {
@@ -512,12 +481,12 @@ def test_rewrite_prompt_sequence_uses_current_rollout_payload_shape():
     }
     assert captured["body"]["messages"][1]["role"] == "user"
     assert prompt_enhancer_module.json.loads(captured["body"]["messages"][1]["content"]) == {
-        "mode": "edit_existing_rollout",
-        "request": (
-            "Rewrite all segment prompts with improved continuity and cinematic detail. "
-            "Keep count and ordering identical."
-        ),
-        "user_instruction": "make it cinematic",
+        "mode":
+        "edit_existing_rollout",
+        "request": ("Rewrite all segment prompts with improved continuity and cinematic detail. "
+                    "Keep count and ordering identical."),
+        "user_instruction":
+        "make it cinematic",
         "current_rollout": {
             "id": "preset_a",
             "label": "Preset A",
@@ -528,11 +497,8 @@ def test_rewrite_prompt_sequence_uses_current_rollout_payload_shape():
 
 def test_rewrite_prompt_sequence_supports_new_rollout_mode():
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"custom_editable","label":"Custom rollout","segment_prompts":['
-            '"A","B","C","D","E","F"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"custom_editable","label":"Custom rollout","segment_prompts":['
+                                   '"A","B","C","D","E","F"]}'))
     captured = {
         "body": None,
     }
@@ -541,10 +507,8 @@ def test_rewrite_prompt_sequence_supports_new_rollout_mode():
         del timeout_seconds
         captured["body"] = body
         return (
-            _chat_payload_with_content(
-                '{"id":"custom_editable","label":"Custom rollout","segment_prompts":['
-                '"A","B","C","D","E","F"]}'
-            ),
+            _chat_payload_with_content('{"id":"custom_editable","label":"Custom rollout","segment_prompts":['
+                                       '"A","B","C","D","E","F"]}'),
             '{"id":"custom_editable","label":"Custom rollout","segment_prompts":['
             '"A","B","C","D","E","F"]}',
         )
@@ -560,30 +524,29 @@ def test_rewrite_prompt_sequence_supports_new_rollout_mode():
             rewrite_model="gpt-test",
             rewrite_temperature=0.2,
             timeout_ms=800,
-        )
-    )
+        ))
 
     assert result.fallback_used is False
     assert result.prompts == ["A", "B", "C", "D", "E", "F"]
     assert prompt_enhancer_module.json.loads(captured["body"]["messages"][1]["content"]) == {
-        "mode": "new_rollout",
-        "request": (
-            "Rewrite all segment prompts with improved continuity and cinematic detail. "
-            "Keep count and ordering identical."
-        ),
-        "user_instruction": "A moonbase corridor thriller with flooding and red alarms",
-        "desired_segment_count": 6,
-        "rollout_id_hint": "custom_editable",
-        "rollout_label_hint": "Custom rollout",
+        "mode":
+        "new_rollout",
+        "request": ("Rewrite all segment prompts with improved continuity and cinematic detail. "
+                    "Keep count and ordering identical."),
+        "user_instruction":
+        "A moonbase corridor thriller with flooding and red alarms",
+        "desired_segment_count":
+        6,
+        "rollout_id_hint":
+        "custom_editable",
+        "rollout_label_hint":
+        "Custom rollout",
     }
 
 
 def test_rewrite_prompt_sequence_uses_session_override_system_prompt():
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     enhancer.rewrite_all_system_prompt = "shared system prompt"
     captured = {
         "body": None,
@@ -593,9 +556,7 @@ def test_rewrite_prompt_sequence_uses_session_override_system_prompt():
         del timeout_seconds
         captured["body"] = body
         return (
-            _chat_payload_with_content(
-                '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-            ),
+            _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'),
             '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}',
         )
 
@@ -609,8 +570,7 @@ def test_rewrite_prompt_sequence_uses_session_override_system_prompt():
             rewrite_instruction="make it cinematic",
             rewrite_model="gpt-test",
             system_prompt_override="session specific system prompt",
-        )
-    )
+        ))
 
     assert result.fallback_used is False
     assert captured["body"]["messages"][0] == {
@@ -621,10 +581,7 @@ def test_rewrite_prompt_sequence_uses_session_override_system_prompt():
 
 def test_resolve_rewrite_new_rollout_system_prompt_uses_dedicated_prompt():
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     enhancer.rewrite_all_system_prompt = "shared rewrite system prompt"
     enhancer.rewrite_user_system_prompt = "new rollout rewrite system prompt"
 
@@ -635,24 +592,17 @@ def test_resolve_rewrite_new_rollout_system_prompt_uses_dedicated_prompt():
 
 def test_resolve_rewrite_new_rollout_system_prompt_prefers_override():
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     enhancer.rewrite_all_system_prompt = "shared rewrite system prompt"
     enhancer.rewrite_user_system_prompt = "new rollout rewrite system prompt"
 
-    resolved = enhancer.resolve_rewrite_new_rollout_system_prompt(
-        "session specific system prompt"
-    )
+    resolved = enhancer.resolve_rewrite_new_rollout_system_prompt("session specific system prompt")
 
     assert resolved == "session specific system prompt"
 
 
 def test_generate_auto_prompt_uses_selected_model():
-    enhancer = _build_test_enhancer(
-        _chat_payload_with_content('{"next_prompt":"Auto next"}')
-    )
+    enhancer = _build_test_enhancer(_chat_payload_with_content('{"next_prompt":"Auto next"}'))
     enhancer.auto_system_prompt = "auto system prompt"
     enhancer.rewrite_model_options = ["gpt-test", "gpt-alt"]
     enhancer.rewrite_default_model = "gpt-test"
@@ -680,8 +630,7 @@ def test_generate_auto_prompt_uses_selected_model():
             next_segment_idx=2,
             model="gpt-alt",
             timeout_ms=800,
-        )
-    )
+        ))
     assert result.fallback_used is False
     assert result.error is None
     assert result.prompt == "Auto next"
@@ -690,9 +639,7 @@ def test_generate_auto_prompt_uses_selected_model():
 
 
 def test_enhance_prompt_uses_selected_model():
-    enhancer = _build_test_enhancer(
-        _chat_payload_with_content('{"next_prompt":"Enhanced next"}')
-    )
+    enhancer = _build_test_enhancer(_chat_payload_with_content('{"next_prompt":"Enhanced next"}'))
     enhancer.enhance_system_prompt = "enhance system prompt"
     enhancer.auto_system_prompt = "auto system prompt"
     enhancer.rewrite_model_options = ["gpt-test", "gpt-alt"]
@@ -722,8 +669,7 @@ def test_enhance_prompt_uses_selected_model():
             next_segment_idx=2,
             model="gpt-alt",
             timeout_ms=800,
-        )
-    )
+        ))
     assert result.fallback_used is False
     assert result.error is None
     assert result.prompt == "Enhanced next"
@@ -732,9 +678,7 @@ def test_enhance_prompt_uses_selected_model():
 
 
 def test_enhance_prompt_single_clip_uses_auto_extension_prompt_and_prompt_field():
-    enhancer = _build_test_enhancer(
-        _chat_payload_with_content('{"prompt":"Extended single clip"}')
-    )
+    enhancer = _build_test_enhancer(_chat_payload_with_content('{"prompt":"Extended single clip"}'))
     enhancer.enhance_system_prompt = "enhance system prompt"
     enhancer.auto_system_prompt = "auto system prompt"
     enhancer.rewrite_model_options = ["gpt-test", "gpt-alt"]
@@ -764,14 +708,12 @@ def test_enhance_prompt_single_clip_uses_auto_extension_prompt_and_prompt_field(
 
     enhancer._request_content = _fake_request_content  # type: ignore[attr-defined]
 
-    result = asyncio.run(
-        enhancer.enhance_prompt(
-            "short 5s idea",
-            mode="single_clip",
-            model="gpt-alt",
-            timeout_ms=800,
-        )
-    )
+    result = asyncio.run(enhancer.enhance_prompt(
+        "short 5s idea",
+        mode="single_clip",
+        model="gpt-alt",
+        timeout_ms=800,
+    ))
     assert result.fallback_used is False
     assert result.error is None
     assert result.prompt == "Extended single clip"
@@ -784,17 +726,15 @@ def test_enhance_prompt_single_clip_uses_auto_extension_prompt_and_prompt_field(
             "single 5-second LTX-2.3 video clip. Respond with "
             'valid JSON only as {"prompt": "..."}.'  # noqa: E501
         ),
-        "user_prompt": "short 5s idea",
+        "user_prompt":
+        "short 5s idea",
     }
 
 
 def test_enhance_prompt_single_clip_rejects_plain_text_response():
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            "Medium shot of a woman by a rainy cafe window as she lifts her "
-            "phone, exhales softly, and the camera makes a slow push in."
-        )
-    )
+        _chat_payload_with_content("Medium shot of a woman by a rainy cafe window as she lifts her "
+                                   "phone, exhales softly, and the camera makes a slow push in."))
     enhancer.auto_system_prompt = "auto system prompt"
 
     result = asyncio.run(
@@ -803,17 +743,14 @@ def test_enhance_prompt_single_clip_rejects_plain_text_response():
             mode="single_clip",
             model="gpt-test",
             timeout_ms=800,
-        )
-    )
+        ))
     assert result.fallback_used is True
     assert "No JSON object found in assistant response." in result.error
     assert result.prompt == ""
 
 
 def test_enhance_prompt_single_clip_rejects_segment_prompts_json():
-    enhancer = _build_test_enhancer(
-        _chat_payload_with_content('{"segment_prompts":["A","B"]}')
-    )
+    enhancer = _build_test_enhancer(_chat_payload_with_content('{"segment_prompts":["A","B"]}'))
     enhancer.auto_system_prompt = "auto system prompt"
     enhancer.rewrite_model_options = ["gpt-test"]
     enhancer.rewrite_default_model = "gpt-test"
@@ -824,17 +761,14 @@ def test_enhance_prompt_single_clip_rejects_segment_prompts_json():
             mode="single_clip",
             model="gpt-test",
             timeout_ms=800,
-        )
-    )
+        ))
     assert result.fallback_used is True
     assert result.prompt == ""
     assert "Missing prompt string." in (result.error or "")
 
 
 def test_enhance_prompt_requires_json_and_does_not_fallback_to_raw_text():
-    enhancer = _build_test_enhancer(
-        _chat_payload_with_content("A cinematic continuation with slow dolly movement.")
-    )
+    enhancer = _build_test_enhancer(_chat_payload_with_content("A cinematic continuation with slow dolly movement."))
     enhancer.enhance_system_prompt = "enhance system prompt"
     enhancer.rewrite_model_options = ["gpt-test"]
     enhancer.rewrite_default_model = "gpt-test"
@@ -846,17 +780,14 @@ def test_enhance_prompt_requires_json_and_does_not_fallback_to_raw_text():
             next_segment_idx=2,
             model="gpt-test",
             timeout_ms=800,
-        )
-    )
+        ))
     assert result.fallback_used is True
     assert result.prompt == ""
     assert "No JSON object found in assistant response." in (result.error or "")
 
 
 def test_generate_auto_prompt_requires_json_and_does_not_fallback_to_raw_text():
-    enhancer = _build_test_enhancer(
-        _chat_payload_with_content("A calm, grounded continuation with subtle motion.")
-    )
+    enhancer = _build_test_enhancer(_chat_payload_with_content("A calm, grounded continuation with subtle motion."))
     enhancer.auto_system_prompt = "auto system prompt"
     enhancer.rewrite_model_options = ["gpt-test"]
     enhancer.rewrite_default_model = "gpt-test"
@@ -867,34 +798,30 @@ def test_generate_auto_prompt_requires_json_and_does_not_fallback_to_raw_text():
             next_segment_idx=2,
             model="gpt-test",
             timeout_ms=800,
-        )
-    )
+        ))
     assert result.fallback_used is True
     assert result.prompt == ""
     assert "No JSON object found in assistant response." in (result.error or "")
 
 
 def test_rewrite_prompt_sequence_includes_raw_json_when_content_empty():
-    enhancer = _build_test_enhancer(
-        {
-            "choices": [
-                {
-                    "finish_reason": "length",
-                    "message": {
-                        "content": [],
-                        "refusal": None,
-                    },
-                }
-            ],
-            "usage": {"completion_tokens": 0},
-        }
-    )
+    enhancer = _build_test_enhancer({
+        "choices": [{
+            "finish_reason": "length",
+            "message": {
+                "content": [],
+                "refusal": None,
+            },
+        }],
+        "usage": {
+            "completion_tokens": 0
+        },
+    })
     result = asyncio.run(
         enhancer.rewrite_prompt_sequence(
             ["prompt one", "prompt two"],
             rewrite_instruction="make it cinematic",
-        )
-    )
+        ))
     assert result.fallback_used is True
     assert "No rewrite segment prompts found in assistant response." in (result.error or "")
     assert isinstance(result.raw_response_text, str)
@@ -903,10 +830,7 @@ def test_rewrite_prompt_sequence_includes_raw_json_when_content_empty():
 
 def test_get_rewrite_model_config_returns_fixed_defaults():
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     enhancer.rewrite_default_model = "gpt-oss-120b"
     enhancer.rewrite_model_options = ["gpt-oss-120b"]
 
@@ -918,10 +842,7 @@ def test_get_rewrite_model_config_returns_fixed_defaults():
 
 def test_get_prompt_config_includes_auto_extension_prompt():
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     enhancer.enhance_system_prompt_path = "/tmp/next.md"
     enhancer.auto_system_prompt_path = "/tmp/auto.md"
     enhancer.rewrite_all_system_prompt_path = "/tmp/rewrite.md"
@@ -948,19 +869,14 @@ def test_get_prompt_config_includes_auto_extension_prompt():
 
 def test_get_prompt_config_reports_loaded_fallback_prompt_path(tmp_path):
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     rewrite_fallback_path = tmp_path / "rewrite_window_system_prompt.md"
     rewrite_fallback_path.write_text("rewrite prompt\n", encoding="utf-8")
     next_path = tmp_path / "next.md"
     next_path.write_text("next prompt\n", encoding="utf-8")
     auto_path = tmp_path / "auto.md"
     auto_path.write_text("auto prompt\n", encoding="utf-8")
-    enhancer.rewrite_all_system_prompt_path = str(
-        tmp_path / "prompts.local" / "rewrite_window_system_prompt.md"
-    )
+    enhancer.rewrite_all_system_prompt_path = str(tmp_path / "prompts.local" / "rewrite_window_system_prompt.md")
     enhancer.rewrite_all_system_prompt_fallback_path = str(rewrite_fallback_path)
     enhancer.enhance_system_prompt_path = str(next_path)
     enhancer.auto_system_prompt_path = str(auto_path)
@@ -973,14 +889,9 @@ def test_get_prompt_config_reports_loaded_fallback_prompt_path(tmp_path):
     assert config["rewrite_window_system_prompt_path"] == str(rewrite_fallback_path)
 
 
-def test_reload_system_prompts_falls_back_to_rewrite_window_when_user_prompt_empty(
-    tmp_path,
-):
+def test_reload_system_prompts_falls_back_to_rewrite_window_when_user_prompt_empty(tmp_path, ):
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     next_path = tmp_path / "next.md"
     auto_path = tmp_path / "auto.md"
     rewrite_path = tmp_path / "rewrite_window_system_prompt.md"
@@ -1007,10 +918,7 @@ def test_reload_system_prompts_falls_back_to_rewrite_window_when_user_prompt_emp
 
 def test_save_prompt_config_updates_auto_extension_prompt(tmp_path):
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     next_path = tmp_path / "next.md"
     auto_path = tmp_path / "auto.md"
     rewrite_path = tmp_path / "rewrite.md"
@@ -1021,9 +929,7 @@ def test_save_prompt_config_updates_auto_extension_prompt(tmp_path):
     enhancer.auto_system_prompt_path = str(auto_path)
     enhancer.rewrite_all_system_prompt_path = str(rewrite_path)
 
-    config = enhancer.save_prompt_config(
-        auto_extension_system_prompt="auto updated",
-    )
+    config = enhancer.save_prompt_config(auto_extension_system_prompt="auto updated", )
 
     assert auto_path.read_text(encoding="utf-8").strip() == "auto updated"
     assert config["auto_extension_system_prompt"] == "auto updated"
@@ -1031,10 +937,7 @@ def test_save_prompt_config_updates_auto_extension_prompt(tmp_path):
 
 def test_save_prompt_config_updates_rewrite_user_prompt(tmp_path):
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     next_path = tmp_path / "next.md"
     auto_path = tmp_path / "auto.md"
     rewrite_path = tmp_path / "rewrite.md"
@@ -1052,9 +955,7 @@ def test_save_prompt_config_updates_rewrite_user_prompt(tmp_path):
     enhancer.rewrite_all_system_prompt_fallback_path = None
     enhancer.rewrite_user_system_prompt_fallback_path = None
 
-    config = enhancer.save_prompt_config(
-        rewrite_user_system_prompt="rewrite user updated",
-    )
+    config = enhancer.save_prompt_config(rewrite_user_system_prompt="rewrite user updated", )
 
     assert rewrite_user_path.read_text(encoding="utf-8").strip() == "rewrite user updated"
     assert config["rewrite_user_system_prompt"] == "rewrite user updated"
@@ -1062,10 +963,7 @@ def test_save_prompt_config_updates_rewrite_user_prompt(tmp_path):
 
 def test_save_prompt_config_updates_rewrite_model(tmp_path):
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     next_path = tmp_path / "next.md"
     auto_path = tmp_path / "auto.md"
     rewrite_path = tmp_path / "rewrite.md"
@@ -1081,9 +979,7 @@ def test_save_prompt_config_updates_rewrite_model(tmp_path):
     enhancer.rewrite_default_model = "gpt-test"
     enhancer.rewrite_model_options = ["gpt-test", "gpt-alt"]
 
-    config = enhancer.save_prompt_config(
-        rewrite_model="gpt-alt",
-    )
+    config = enhancer.save_prompt_config(rewrite_model="gpt-alt", )
 
     assert enhancer.rewrite_default_model == "gpt-alt"
     assert config["rewrite_model"] == "gpt-alt"
@@ -1092,10 +988,7 @@ def test_save_prompt_config_updates_rewrite_model(tmp_path):
 
 def test_save_prompt_config_updates_rewrite_temperature(tmp_path):
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     next_path = tmp_path / "next.md"
     auto_path = tmp_path / "auto.md"
     rewrite_path = tmp_path / "rewrite.md"
@@ -1109,9 +1002,7 @@ def test_save_prompt_config_updates_rewrite_temperature(tmp_path):
     enhancer.auto_system_prompt_fallback_path = None
     enhancer.rewrite_all_system_prompt_fallback_path = None
 
-    config = enhancer.save_prompt_config(
-        rewrite_temperature=1.3,
-    )
+    config = enhancer.save_prompt_config(rewrite_temperature=1.3, )
 
     assert enhancer.rewrite_default_temperature == 1.3
     assert config["rewrite_temperature"] == 1.3
@@ -1119,10 +1010,7 @@ def test_save_prompt_config_updates_rewrite_temperature(tmp_path):
 
 def test_save_prompt_config_creates_versioned_backup_for_existing_prompt(tmp_path):
     enhancer = _build_test_enhancer(
-        _chat_payload_with_content(
-            '{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'
-        )
-    )
+        _chat_payload_with_content('{"id":"preset_a","label":"Preset A","segment_prompts":["A","B"]}'))
     next_path = tmp_path / "next.md"
     auto_path = tmp_path / "auto.md"
     rewrite_path = tmp_path / "rewrite_window_system_prompt.md"
@@ -1136,13 +1024,9 @@ def test_save_prompt_config_creates_versioned_backup_for_existing_prompt(tmp_pat
     enhancer.auto_system_prompt_fallback_path = None
     enhancer.rewrite_all_system_prompt_fallback_path = None
 
-    enhancer.save_prompt_config(
-        rewrite_window_system_prompt="rewrite updated",
-    )
+    enhancer.save_prompt_config(rewrite_window_system_prompt="rewrite updated", )
 
-    backup_paths = sorted(
-        tmp_path.glob("rewrite_window_system_prompt.*.bak.md")
-    )
+    backup_paths = sorted(tmp_path.glob("rewrite_window_system_prompt.*.bak.md"))
 
     assert rewrite_path.read_text(encoding="utf-8").strip() == "rewrite updated"
     assert len(backup_paths) == 1

apps/dreamverse/dreamverse/tests/test_realtime_stress.py

@@ -27,13 +27,8 @@ try:
 except ModuleNotFoundError:
     websockets = None  # type: ignore[assignment]
 
-
-DEFAULT_PRESET_FILE = (
-    Path(__file__).resolve().parents[2]
-    / "web"
-    / "prompts"
-    / "selected_ltx2_continuation_story_presets.json"
-)
+DEFAULT_PRESET_FILE = (Path(__file__).resolve().parents[2] / "web" / "prompts" /
+                       "selected_ltx2_continuation_story_presets.json")
 
 
 def utc_now_iso() -> str:
@@ -65,10 +60,7 @@ def safe_percentile(values: list[float], percentile: float) -> float | None:
     if lower == upper:
         return sorted_values[lower]
     fraction = rank - lower
-    return (
-        sorted_values[lower]
-        + (sorted_values[upper] - sorted_values[lower]) * fraction
-    )
+    return (sorted_values[lower] + (sorted_values[upper] - sorted_values[lower]) * fraction)
 
 
 def summarize_series(values: list[float]) -> dict[str, float | int | None]:
@@ -145,24 +137,16 @@ def load_curated_prompts(
     selected_id = str(selected.get("id", "")).strip() or "unknown_preset"
     raw_prompts = selected.get("segment_prompts", [])
     if not isinstance(raw_prompts, list):
-        raise ValueError(
-            f"Preset {selected_id} has invalid segment_prompts (must be list)."
-        )
+        raise ValueError(f"Preset {selected_id} has invalid segment_prompts (must be list).")
 
-    prompts = [
-        str(prompt).strip()
-        for prompt in raw_prompts
-        if isinstance(prompt, str) and str(prompt).strip()
-    ]
+    prompts = [str(prompt).strip() for prompt in raw_prompts if isinstance(prompt, str) and str(prompt).strip()]
     if not prompts:
         raise ValueError(f"Preset {selected_id} has no non-empty prompts.")
 
     limited = prompts[:curated_limit]
     if not limited:
-        raise ValueError(
-            f"curated_limit={curated_limit} produced no prompts for preset "
-            f"{selected_id}."
-        )
+        raise ValueError(f"curated_limit={curated_limit} produced no prompts for preset "
+                         f"{selected_id}.")
     return selected_id, limited, len(prompts)
 
 
@@ -224,11 +208,11 @@ async def run_single_session(
 
     try:
         async with websockets.connect(
-            url,
-            max_size=None,
-            ping_interval=None,
-            open_timeout=connect_timeout_s,
-            close_timeout=2.0,
+                url,
+                max_size=None,
+                ping_interval=None,
+                open_timeout=connect_timeout_s,
+                close_timeout=2.0,
         ) as ws:
             connect_finish_monotonic = time.monotonic()
             session_data["connect_finish_ts_utc"] = utc_now_iso()
@@ -249,9 +233,7 @@ async def run_single_session(
                 timeout_remaining = session_timeout_s - elapsed_s
                 if timeout_remaining <= 0:
                     session_data["status"] = "timeout"
-                    session_data["error"] = (
-                        f"Session timed out after {session_timeout_s:.1f}s."
-                    )
+                    session_data["error"] = (f"Session timed out after {session_timeout_s:.1f}s.")
                     break
 
                 recv_start_epoch = time.time()
@@ -265,9 +247,7 @@ async def run_single_session(
                     )
                 except asyncio.TimeoutError:
                     session_data["status"] = "timeout"
-                    session_data["error"] = (
-                        "Timed out waiting for websocket message."
-                    )
+                    session_data["error"] = ("Timed out waiting for websocket message.")
                     break
                 except Exception as exc:
                     session_data["status"] = "failed"
@@ -288,20 +268,16 @@ async def run_single_session(
 
                     chunk_gap_ms: float | None = None
                     if last_chunk_finish_monotonic is not None:
-                        chunk_gap_ms = (
-                            recv_finish_monotonic - last_chunk_finish_monotonic
-                        ) * 1000.0
+                        chunk_gap_ms = (recv_finish_monotonic - last_chunk_finish_monotonic) * 1000.0
 
-                    session_data["chunks"].append(
-                        {
-                            "segment_idx": current_segment_idx,
-                            "chunk_idx": session_data["total_chunks"],
-                            "size_bytes": len(message),
-                            "chunk_start_ts_utc": recv_start_iso,
-                            "chunk_finish_ts_utc": recv_finish_iso,
-                            "chunk_gap_ms": chunk_gap_ms,
-                        }
-                    )
+                    session_data["chunks"].append({
+                        "segment_idx": current_segment_idx,
+                        "chunk_idx": session_data["total_chunks"],
+                        "size_bytes": len(message),
+                        "chunk_start_ts_utc": recv_start_iso,
+                        "chunk_finish_ts_utc": recv_finish_iso,
+                        "chunk_gap_ms": chunk_gap_ms,
+                    })
                     last_chunk_finish_monotonic = recv_finish_monotonic
                     last_chunk_finish_epoch = recv_finish_epoch
                     session_data["last_chunk_finish_ts_utc"] = recv_finish_iso
@@ -321,9 +297,7 @@ async def run_single_session(
                 if msg_type == "gpu_assigned":
                     session_data["gpu_assigned_ts_utc"] = recv_finish_iso
                     if connect_finish_monotonic is not None:
-                        session_data["queue_wait_ms"] = (
-                            recv_finish_monotonic - connect_finish_monotonic
-                        ) * 1000.0
+                        session_data["queue_wait_ms"] = (recv_finish_monotonic - connect_finish_monotonic) * 1000.0
                 elif msg_type == "ltx2_stream_start":
                     if initial_total_segments is None:
                         parsed_total = parse_int(data.get("total_segments"))
@@ -338,20 +312,13 @@ async def run_single_session(
                     session_data["media_segments_completed"] += 1
                     if first_media_segment_complete_epoch is None:
                         first_media_segment_complete_epoch = recv_finish_epoch
-                        session_data[
-                            "first_media_segment_complete_ts_utc"
-                        ] = recv_finish_iso
+                        session_data["first_media_segment_complete_ts_utc"] = recv_finish_iso
                 elif msg_type == "ltx2_segment_complete":
                     session_data["segments_completed"] += 1
                     seg_idx = parse_int(data.get("segment_idx"))
-                    if (
-                        initial_total_segments is not None
-                        and seg_idx is not None
-                        and seg_idx >= initial_total_segments
-                    ):
-                        session_data[
-                            "target_segment_complete_ts_utc"
-                        ] = recv_finish_iso
+                    if (initial_total_segments is not None and seg_idx is not None
+                            and seg_idx >= initial_total_segments):
+                        session_data["target_segment_complete_ts_utc"] = recv_finish_iso
                         await asyncio.sleep(post_complete_wait_s)
                         session_data["leave_sent_ts_utc"] = utc_now_iso()
                         try:
@@ -362,15 +329,11 @@ async def run_single_session(
                         break
                 elif msg_type == "session_timeout":
                     session_data["status"] = "timeout"
-                    session_data["error"] = str(
-                        data.get("message") or "Backend session timeout"
-                    )
+                    session_data["error"] = str(data.get("message") or "Backend session timeout")
                     break
                 elif msg_type == "error":
                     session_data["status"] = "failed"
-                    session_data["error"] = str(
-                        data.get("message") or "Backend error message"
-                    )
+                    session_data["error"] = str(data.get("message") or "Backend error message")
                     break
 
             if session_data["status"] == "failed" and session_data["error"] is None:
@@ -379,29 +342,18 @@ async def run_single_session(
         session_data["status"] = "failed"
         session_data["error"] = f"WebSocket connect/run failed: {exc}"
 
-    if (
-        first_chunk_finish_epoch is not None
-        and last_chunk_finish_epoch is not None
-        and session_data["total_chunk_bytes"] > 0
-    ):
+    if (first_chunk_finish_epoch is not None and last_chunk_finish_epoch is not None
+            and session_data["total_chunk_bytes"] > 0):
         duration_s = last_chunk_finish_epoch - first_chunk_finish_epoch
         if duration_s > 0:
-            session_data["session_goodput_mbps"] = (
-                session_data["total_chunk_bytes"] * 8.0 / duration_s / 1_000_000.0
-            )
+            session_data["session_goodput_mbps"] = (session_data["total_chunk_bytes"] * 8.0 / duration_s / 1_000_000.0)
 
-    if (
-        first_chunk_finish_epoch is not None
-        and first_media_segment_complete_epoch is not None
-    ):
-        session_data["first_chunk_before_first_media_complete"] = (
-            first_chunk_finish_epoch < first_media_segment_complete_epoch
-        )
+    if (first_chunk_finish_epoch is not None and first_media_segment_complete_epoch is not None):
+        session_data["first_chunk_before_first_media_complete"] = (first_chunk_finish_epoch
+                                                                   < first_media_segment_complete_epoch)
 
     session_data["close_ts_utc"] = utc_now_iso()
-    session_data["duration_ms"] = (
-        time.monotonic() - session_start_monotonic
-    ) * 1000.0
+    session_data["duration_ms"] = (time.monotonic() - session_start_monotonic) * 1000.0
     return session_data
 
 
@@ -412,14 +364,11 @@ async def run_worker_sessions(
     config: dict[str, Any],
 ) -> list[dict[str, Any]]:
     tasks = [
-        asyncio.create_task(
-            run_single_session(
-                worker_id=worker_id,
-                worker_session_idx=idx,
-                config=config,
-            )
-        )
-        for idx in range(session_count)
+        asyncio.create_task(run_single_session(
+            worker_id=worker_id,
+            worker_session_idx=idx,
+            config=config,
+        )) for idx in range(session_count)
     ]
     if not tasks:
         return []
@@ -437,29 +386,23 @@ def worker_entry(
     try:
         ready_queue.put({"worker_id": worker_id, "status": "ready"})
         start_event.wait()
-        sessions = asyncio.run(
-            run_worker_sessions(
-                worker_id=worker_id,
-                session_count=session_count,
-                config=config,
-            )
-        )
-        result_queue.put(
-            {
-                "worker_id": worker_id,
-                "status": "ok",
-                "sessions": sessions,
-            }
-        )
+        sessions = asyncio.run(run_worker_sessions(
+            worker_id=worker_id,
+            session_count=session_count,
+            config=config,
+        ))
+        result_queue.put({
+            "worker_id": worker_id,
+            "status": "ok",
+            "sessions": sessions,
+        })
     except Exception as exc:
-        result_queue.put(
-            {
-                "worker_id": worker_id,
-                "status": "error",
-                "error": str(exc),
-                "traceback": traceback.format_exc(),
-            }
-        )
+        result_queue.put({
+            "worker_id": worker_id,
+            "status": "error",
+            "error": str(exc),
+            "traceback": traceback.format_exc(),
+        })
 
 
 def build_summary(
@@ -517,33 +460,22 @@ def build_summary(
     if len(all_chunk_finish_epochs) >= 2 and total_chunk_bytes > 0:
         duration_s = max(all_chunk_finish_epochs) - min(all_chunk_finish_epochs)
         if duration_s > 0:
-            global_goodput_mbps = (
-                total_chunk_bytes * 8.0 / duration_s / 1_000_000.0
-            )
+            global_goodput_mbps = (total_chunk_bytes * 8.0 / duration_s / 1_000_000.0)
 
-    bucket_throughputs_mbps = [
-        (bytes_count * 8.0) / 1_000_000.0
-        for _, bytes_count in sorted(bucket_bytes.items())
-    ]
+    bucket_throughputs_mbps = [(bytes_count * 8.0) / 1_000_000.0 for _, bytes_count in sorted(bucket_bytes.items())]
     bucket_stats = summarize_series(bucket_throughputs_mbps)
 
-    chunk_gap_threshold_breaches = [
-        value for value in chunk_gaps if value >= chunk_gap_threshold_ms
-    ]
+    chunk_gap_threshold_breaches = [value for value in chunk_gaps if value >= chunk_gap_threshold_ms]
     non_success = len(sessions) - status_counts.get("success", 0)
 
     fail_reasons: list[str] = []
     if non_success > 0:
-        fail_reasons.append(
-            f"{non_success} session(s) did not complete successfully."
-        )
+        fail_reasons.append(f"{non_success} session(s) did not complete successfully.")
     if not chunk_gaps:
         fail_reasons.append("No chunk gap data collected.")
     if chunk_gap_threshold_breaches:
-        fail_reasons.append(
-            f"{len(chunk_gap_threshold_breaches)} chunk gap(s) were >= "
-            f"{chunk_gap_threshold_ms:.0f}ms."
-        )
+        fail_reasons.append(f"{len(chunk_gap_threshold_breaches)} chunk gap(s) were >= "
+                            f"{chunk_gap_threshold_ms:.0f}ms.")
 
     passed = len(fail_reasons) == 0
     progressive_ratio = None
@@ -554,20 +486,18 @@ def build_summary(
         "passed": passed,
         "fail_reasons": fail_reasons,
         "sessions": {
-            "total": len(sessions),
-            "success": status_counts.get("success", 0),
-            "failed": status_counts.get("failed", 0),
-            "timeout": status_counts.get("timeout", 0),
-            "protocol_error": status_counts.get("protocol_error", 0),
-            "other": (
-                len(sessions)
-                - (
-                    status_counts.get("success", 0)
-                    + status_counts.get("failed", 0)
-                    + status_counts.get("timeout", 0)
-                    + status_counts.get("protocol_error", 0)
-                )
-            ),
+            "total":
+            len(sessions),
+            "success":
+            status_counts.get("success", 0),
+            "failed":
+            status_counts.get("failed", 0),
+            "timeout":
+            status_counts.get("timeout", 0),
+            "protocol_error":
+            status_counts.get("protocol_error", 0),
+            "other": (len(sessions) - (status_counts.get("success", 0) + status_counts.get("failed", 0) +
+                                       status_counts.get("timeout", 0) + status_counts.get("protocol_error", 0))),
         },
         "chunk_gap_ms": {
             **chunk_gap_stats,
@@ -606,51 +536,39 @@ def print_summary(
     bucket_bw = bandwidth["bucketed_1s"]
 
     print("=== LTX2 Realtime Stress Test Summary ===")
-    print(
-        "Run: "
-        f"url={run_info['url']} clients={run_info['clients']} "
-        f"processes={run_info['processes']} "
-        f"preset={run_info['preset_id']} "
-        f"curated_limit={run_info['curated_limit']}"
-    )
-    print(
-        "Sessions: "
-        f"total={sessions['total']} success={sessions['success']} "
-        f"failed={sessions['failed']} timeout={sessions['timeout']} "
-        f"protocol_error={sessions['protocol_error']}"
-    )
-    print(
-        "Chunk gap ms: "
-        f"min={format_num(chunk_gap['min'])} "
-        f"p50={format_num(chunk_gap['p50'])} "
-        f"p95={format_num(chunk_gap['p95'])} "
-        f"p99={format_num(chunk_gap['p99'])} "
-        f"max={format_num(chunk_gap['max'])} "
-        f"threshold={format_num(chunk_gap['threshold_ms'])} "
-        f"breaches={chunk_gap['breach_count']}"
-    )
-    print(
-        "Queue wait ms: "
-        f"min={format_num(queue_wait['min'])} "
-        f"p50={format_num(queue_wait['p50'])} "
-        f"p95={format_num(queue_wait['p95'])} "
-        f"max={format_num(queue_wait['max'])}"
-    )
+    print("Run: "
+          f"url={run_info['url']} clients={run_info['clients']} "
+          f"processes={run_info['processes']} "
+          f"preset={run_info['preset_id']} "
+          f"curated_limit={run_info['curated_limit']}")
+    print("Sessions: "
+          f"total={sessions['total']} success={sessions['success']} "
+          f"failed={sessions['failed']} timeout={sessions['timeout']} "
+          f"protocol_error={sessions['protocol_error']}")
+    print("Chunk gap ms: "
+          f"min={format_num(chunk_gap['min'])} "
+          f"p50={format_num(chunk_gap['p50'])} "
+          f"p95={format_num(chunk_gap['p95'])} "
+          f"p99={format_num(chunk_gap['p99'])} "
+          f"max={format_num(chunk_gap['max'])} "
+          f"threshold={format_num(chunk_gap['threshold_ms'])} "
+          f"breaches={chunk_gap['breach_count']}")
+    print("Queue wait ms: "
+          f"min={format_num(queue_wait['min'])} "
+          f"p50={format_num(queue_wait['p50'])} "
+          f"p95={format_num(queue_wait['p95'])} "
+          f"max={format_num(queue_wait['max'])}")
     ratio = progressive["ratio"]
     ratio_text = "n/a" if ratio is None else f"{ratio * 100:.2f}%"
-    print(
-        "Progressive streaming: "
-        f"{progressive['success_sessions']}/"
-        f"{progressive['eligible_sessions']} ({ratio_text})"
-    )
-    print(
-        "Bandwidth Mbps: "
-        f"per_session_avg={format_num(per_session_bw['avg'])} "
-        f"per_session_p95={format_num(per_session_bw['p95'])} "
-        f"global={format_num(bandwidth['global_goodput_mbps'])} "
-        f"bucket_avg={format_num(bucket_bw['avg_mbps'])} "
-        f"bucket_peak={format_num(bucket_bw['peak_mbps'])}"
-    )
+    print("Progressive streaming: "
+          f"{progressive['success_sessions']}/"
+          f"{progressive['eligible_sessions']} ({ratio_text})")
+    print("Bandwidth Mbps: "
+          f"per_session_avg={format_num(per_session_bw['avg'])} "
+          f"per_session_p95={format_num(per_session_bw['p95'])} "
+          f"global={format_num(bandwidth['global_goodput_mbps'])} "
+          f"bucket_avg={format_num(bucket_bw['avg_mbps'])} "
+          f"bucket_peak={format_num(bucket_bw['peak_mbps'])}")
     print(f"VERDICT: {'PASS' if summary['passed'] else 'FAIL'}")
     if summary["fail_reasons"]:
         print("Fail reasons:")
@@ -670,10 +588,8 @@ def distribute_sessions(total_clients: int, process_count: int) -> list[int]:
 
 def run_stress(args: argparse.Namespace) -> tuple[dict[str, Any], int]:
     if websockets is None:
-        raise RuntimeError(
-            "Missing dependency: websockets. Install it before running this "
-            "stress test."
-        )
+        raise RuntimeError("Missing dependency: websockets. Install it before running this "
+                           "stress test.")
 
     preset_file = Path(args.preset_file).expanduser().resolve()
     selected_preset_id, curated_prompts, total_prompt_count = load_curated_prompts(
@@ -735,13 +651,8 @@ def run_stress(args: argparse.Namespace) -> tuple[dict[str, Any], int]:
 
         start_event.set()
 
-        result_deadline = (
-            time.monotonic()
-            + args.connect_timeout_s
-            + args.session_timeout_s
-            + args.post_complete_wait_s
-            + 180.0
-        )
+        result_deadline = (time.monotonic() + args.connect_timeout_s + args.session_timeout_s +
+                           args.post_complete_wait_s + 180.0)
         worker_results: list[dict[str, Any]] = []
         while len(worker_results) < len(processes):
             timeout_s = max(0.1, result_deadline - time.monotonic())
@@ -765,24 +676,20 @@ def run_stress(args: argparse.Namespace) -> tuple[dict[str, Any], int]:
             if result.get("status") == "ok":
                 sessions.extend(result.get("sessions", []))
             else:
-                worker_errors.append(
-                    {
-                        "worker_id": result.get("worker_id"),
-                        "error": result.get("error"),
-                        "traceback": result.get("traceback"),
-                    }
-                )
+                worker_errors.append({
+                    "worker_id": result.get("worker_id"),
+                    "error": result.get("error"),
+                    "traceback": result.get("traceback"),
+                })
 
         received_workers = {result.get("worker_id") for result in worker_results}
         expected_workers = set(range(len(processes)))
         missing_workers = sorted(expected_workers - received_workers)
         for worker_id in missing_workers:
-            worker_errors.append(
-                {
-                    "worker_id": worker_id,
-                    "error": "No worker result received.",
-                }
-            )
+            worker_errors.append({
+                "worker_id": worker_id,
+                "error": "No worker result received.",
+            })
 
         run_end_epoch = time.time()
         run_end_iso = iso_from_epoch(run_end_epoch)
@@ -795,9 +702,8 @@ def run_stress(args: argparse.Namespace) -> tuple[dict[str, Any], int]:
 
         if worker_errors:
             summary["passed"] = False
-            summary["fail_reasons"] = list(summary["fail_reasons"]) + [
-                f"{len(worker_errors)} worker error(s) occurred."
-            ]
+            summary["fail_reasons"] = list(
+                summary["fail_reasons"]) + [f"{len(worker_errors)} worker error(s) occurred."]
 
         output_payload = {
             "run_info": {
@@ -833,9 +739,7 @@ def run_stress(args: argparse.Namespace) -> tuple[dict[str, Any], int]:
 
 
 def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(
-        description="Multiprocess realtime stress test for LTX2 streaming.",
-    )
+    parser = argparse.ArgumentParser(description="Multiprocess realtime stress test for LTX2 streaming.", )
     parser.add_argument(
         "-u",
         "--url",

apps/dreamverse/dreamverse/tests/test_session_init_image.py

@@ -47,13 +47,11 @@ def test_persist_session_init_image_returns_none_when_missing_data():
 
 def test_persist_session_init_image_rejects_unsupported_mime():
     with pytest.raises(ValueError, match="PNG, JPEG, or WebP"):
-        persist_session_init_image(
-            {
-                "name": "frame.gif",
-                "mime_type": "image/gif",
-                "data_url": "data:image/gif;base64,R0lGODlhAQABAAAAACw=",
-            }
-        )
+        persist_session_init_image({
+            "name": "frame.gif",
+            "mime_type": "image/gif",
+            "data_url": "data:image/gif;base64,R0lGODlhAQABAAAAACw=",
+        })
 
 
 def test_persist_session_init_image_rejects_large_payload(monkeypatch):
@@ -66,10 +64,8 @@ def test_persist_session_init_image_rejects_large_payload(monkeypatch):
     monkeypatch.setattr(base64, "b64decode", fake_b64decode)
 
     with pytest.raises(ValueError, match="15 MB or smaller"):
-        persist_session_init_image(
-            {
-                "name": "frame.png",
-                "mime_type": "image/png",
-                "data_url": data_url,
-            }
-        )
+        persist_session_init_image({
+            "name": "frame.png",
+            "mime_type": "image/png",
+            "data_url": data_url,
+        })

apps/dreamverse/gpu-pool.drawio

@@ -70,7 +70,7 @@
                 <mxCell id="dispatcher" value="command dispatcher&#xa;&#xa;gpu_worker_process() branches on&#xa;CommandType; asserts payload type&#xa;&#xa;INIT / WARMUP / RELOAD_MODEL&#xa;USER_JOIN / USER_STEP / USER_LEAVE&#xa;SHUTDOWN" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffe6cc;strokeColor=#d79b00;fontSize=11;align=left;spacingLeft=10;spacingTop=8;fontStyle=1;" parent="1" vertex="1">
                     <mxGeometry x="120" y="1120" width="240" height="120" as="geometry"/>
                 </mxCell>
-                <mxCell id="do_step" value="VideoGenerationWorker.generate_step()&#xa;video_generation.py:380&#xa;&#xa;reads + updates ContinuationState,&#xa;calls generator" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#e1d5e7;strokeColor=#9673a6;fontSize=11;align=left;spacingLeft=10;spacingTop=8;fontStyle=1;" parent="1" vertex="1">
+                <mxCell id="do_step" value="VideoGenerationWorker.generate_step()&#xa;ltx2_generation.py:380&#xa;&#xa;reads + updates ContinuationState,&#xa;calls generator" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#e1d5e7;strokeColor=#9673a6;fontSize=11;align=left;spacingLeft=10;spacingTop=8;fontStyle=1;" parent="1" vertex="1">
                     <mxGeometry x="460" y="1120" width="240" height="120" as="geometry"/>
                 </mxCell>
                 <mxCell id="stream_av" value="stream_fmp4()&#xa;av_streaming.py:121&#xa;&#xa;trims overlap, pipes to ffmpeg,&#xa;publishes StreamInit / StreamChunk /&#xa;StreamComplete via callback" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#b1d8d7;strokeColor=#23445d;fontSize=11;align=left;spacingLeft=10;spacingTop=8;fontStyle=1;" parent="1" vertex="1">
@@ -79,13 +79,13 @@
                 <mxCell id="Ot8BU52QTIb4EhyRSe7I-2" value="" style="edgeStyle=none;html=1;" parent="1" source="generator" target="Ot8BU52QTIb4EhyRSe7I-1" edge="1">
                     <mxGeometry relative="1" as="geometry"/>
                 </mxCell>
-                <mxCell id="generator" value="VideoGenerator  (fastvideo)&#xa;&#xa;LTX2 DiT + refine upsampler&#xa;FP4 quant, torch.compile&#xa;&#xa;owned by VideoGenerationWorker&#xa;video_generation.py:211" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#e1d5e7;strokeColor=#9673a6;fontSize=11;" parent="1" vertex="1">
+                <mxCell id="generator" value="VideoGenerator  (fastvideo)&#xa;&#xa;LTX2 DiT + refine upsampler&#xa;FP4 quant, torch.compile&#xa;&#xa;owned by VideoGenerationWorker&#xa;ltx2_generation.py:211" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#e1d5e7;strokeColor=#9673a6;fontSize=11;" parent="1" vertex="1">
                     <mxGeometry x="460" y="1300" width="240" height="100" as="geometry"/>
                 </mxCell>
                 <mxCell id="ffmpeg" value="ffmpeg subprocess&#xa;&#xa;libx264 / *_nvenc&#xa;fragmented mp4" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffffff;strokeColor=#d79b00;fontSize=11;" parent="1" vertex="1">
                     <mxGeometry x="800" y="1300" width="260" height="100" as="geometry"/>
                 </mxCell>
-                <mxCell id="caches" value="ContinuationState&#xa;video_generation.py:89&#xa;&#xa;• video_images: list[PIL.Image]&#xa;• audio_latents: torch.Tensor (CPU)&#xa;&#xa;carried across segments" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#e1d5e7;strokeColor=#9673a6;fontSize=11;align=left;spacingLeft=10;spacingTop=8;" parent="1" vertex="1">
+                <mxCell id="caches" value="ContinuationState&#xa;ltx2_generation.py:89&#xa;&#xa;• video_images: list[PIL.Image]&#xa;• audio_latents: torch.Tensor (CPU)&#xa;&#xa;carried across segments" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#e1d5e7;strokeColor=#9673a6;fontSize=11;align=left;spacingLeft=10;spacingTop=8;" parent="1" vertex="1">
                     <mxGeometry x="120" y="1300" width="240" height="100" as="geometry"/>
                 </mxCell>
                 <mxCell id="e_cp" value="acquire" style="edgeStyle=orthogonalEdgeStyle;rounded=0;html=1;strokeColor=#6c8ebf;endArrow=classic;fontSize=11;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0.5;entryY=0;entryDx=0;entryDy=0;" parent="1" source="client" target="pool" edge="1">
@@ -250,7 +250,7 @@
                         <mxPoint x="690" y="880"/>
                     </Array>
                 </mxCell>
-                <mxCell id="legend" value="Legend&#xa;&#xa;■ blue  client / external&#xa;■ green  main-process pool/slot&#xa;    (methods — italic label)&#xa;■ yellow  containers (routing state)&#xa;■ red  IPC primitives (mp.Queue, mp.RawArray)&#xa;&#xa;Worker subprocess modules:&#xa;■ orange  gpu_pool.py (dispatcher)&#xa;■ lavender  video_generation.py&#xa;■ teal  av_streaming.py&#xa;■ gray  worker_ipc.py (shared types)&#xa;&#xa;Flow:&#xa;  client → pool → slot&#xa;  → _send_command(_tagged) → command_queue&#xa;  → dispatcher → generate_step()&#xa;  → stream_fmp4() → ffmpeg&#xa;  → shared_buf + response_queue&#xa;  → _response_reader → futures / stream_queues&#xa;  → client awaits (via main.py AV loop)" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#f5f5f5;strokeColor=#999999;fontSize=11;align=left;spacingLeft=10;spacingTop=8;" parent="1" vertex="1">
+                <mxCell id="legend" value="Legend&#xa;&#xa;■ blue  client / external&#xa;■ green  main-process pool/slot&#xa;    (methods — italic label)&#xa;■ yellow  containers (routing state)&#xa;■ red  IPC primitives (mp.Queue, mp.RawArray)&#xa;&#xa;Worker subprocess modules:&#xa;■ orange  gpu_pool.py (dispatcher)&#xa;■ lavender  ltx2_generation.py&#xa;■ teal  av_streaming.py&#xa;■ gray  worker_ipc.py (shared types)&#xa;&#xa;Flow:&#xa;  client → pool → slot&#xa;  → _send_command(_tagged) → command_queue&#xa;  → dispatcher → generate_step()&#xa;  → stream_fmp4() → ffmpeg&#xa;  → shared_buf + response_queue&#xa;  → _response_reader → futures / stream_queues&#xa;  → client awaits (via main.py AV loop)" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#f5f5f5;strokeColor=#999999;fontSize=11;align=left;spacingLeft=10;spacingTop=8;" parent="1" vertex="1">
                     <mxGeometry x="39" y="-200" width="270" height="380" as="geometry"/>
                 </mxCell>
                 <mxCell id="Ot8BU52QTIb4EhyRSe7I-1" value="FastVideo video_generator" style="whiteSpace=wrap;html=1;fontSize=11;fillColor=#e1d5e7;strokeColor=#9673a6;rounded=1;" parent="1" vertex="1">
@@ -389,10 +389,10 @@
                 <mxCell id="cw2" value="from fastvideo.entrypoints.video_generator import VideoGenerator&#xa;from fastvideo.models.dits.ltx2 import DEFAULT_LTX2_AUDIO_*&#xa;&#xa;** Dreamverse reaches into fastvideo internals here **" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffe0b2;strokeColor=#d79b00;fontSize=10;align=left;spacingLeft=8;fontFamily=monospace;" parent="1" vertex="1">
                     <mxGeometry x="675" y="695" width="550" height="60" as="geometry"/>
                 </mxCell>
-                <mxCell id="cw3" value="on Command(INIT):&#xa;   VideoGenerationWorker.initialize()    (video_generation.py:247)&#xa;      maybe_download_model(model_id)&#xa;      VideoGenerator.from_pretrained(path, FP4Config, PipelineConfig)&#xa;      load audio VAE, resolve refine upsampler&#xa;      resp_q.put(InitAck(success=True))" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffffff;strokeColor=#d79b00;fontSize=10;align=left;spacingLeft=8;fontFamily=monospace;" parent="1" vertex="1">
+                <mxCell id="cw3" value="on Command(INIT):&#xa;   VideoGenerationWorker.initialize()    (ltx2_generation.py:247)&#xa;      maybe_download_model(model_id)&#xa;      VideoGenerator.from_pretrained(path, FP4Config, PipelineConfig)&#xa;      load audio VAE, resolve refine upsampler&#xa;      resp_q.put(InitAck(success=True))" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffffff;strokeColor=#d79b00;fontSize=10;align=left;spacingLeft=8;fontFamily=monospace;" parent="1" vertex="1">
                     <mxGeometry x="675" y="765" width="550" height="95" as="geometry"/>
                 </mxCell>
-                <mxCell id="cw4" value="on Command(WARMUP) with WarmupPayload:&#xa;   VideoGenerationWorker.warmup(payload.prompt)    (video_generation.py:518)&#xa;      two synthetic segments prime caches + torch.compile&#xa;   resp_q.put(WarmupComplete(timings=...))" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffffff;strokeColor=#d79b00;fontSize=10;align=left;spacingLeft=8;fontFamily=monospace;" parent="1" vertex="1">
+                <mxCell id="cw4" value="on Command(WARMUP) with WarmupPayload:&#xa;   VideoGenerationWorker.warmup(payload.prompt)    (ltx2_generation.py:518)&#xa;      two synthetic segments prime caches + torch.compile&#xa;   resp_q.put(WarmupComplete(timings=...))" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffffff;strokeColor=#d79b00;fontSize=10;align=left;spacingLeft=8;fontFamily=monospace;" parent="1" vertex="1">
                     <mxGeometry x="675" y="870" width="550" height="55" as="geometry"/>
                 </mxCell>
                 <mxCell id="cw5" value="enter main worker loop → waits for JOIN_USER / USER_STEP / LEAVE" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#c8e6c9;strokeColor=#388e3c;fontSize=11;fontStyle=1;fontFamily=monospace;" parent="1" vertex="1">
@@ -534,7 +534,7 @@
                         <mxPoint x="1040" y="1610" as="targetPoint"/>
                     </mxGeometry>
                 </mxCell>
-                <mxCell id="dm11a" value="10a. worker runs:&#xa;VideoGenerationWorker.generate_step()&#xa;    (video_generation.py:380)&#xa;  → generator.generate_video()&#xa;  → updates ContinuationState&#xa;then stream_fmp4() (av_streaming.py:121)&#xa;  → ffmpeg (rawvideo+wav → fmp4)" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffe0b2;strokeColor=#d79b00;fontSize=10;align=left;spacingLeft=8;fontFamily=monospace;" parent="1" vertex="1">
+                <mxCell id="dm11a" value="10a. worker runs:&#xa;VideoGenerationWorker.generate_step()&#xa;    (ltx2_generation.py:380)&#xa;  → generator.generate_video()&#xa;  → updates ContinuationState&#xa;then stream_fmp4() (av_streaming.py:121)&#xa;  → ffmpeg (rawvideo+wav → fmp4)" style="rounded=1;whiteSpace=wrap;html=1;fillColor=#ffe0b2;strokeColor=#d79b00;fontSize=10;align=left;spacingLeft=8;fontFamily=monospace;" parent="1" vertex="1">
                     <mxGeometry x="955" y="1640" width="180" height="70" as="geometry"/>
                 </mxCell>
                 <mxCell id="dm11" value="10b. resp_q.put(MediaInit / MediaChunk / MediaComplete / StepComplete)" style="endArrow=classic;html=1;strokeColor=#b85450;fontSize=10;labelBackgroundColor=#ffffff;" parent="1" edge="1">

apps/dreamverse/scripts/launch/launch_backend_dreamverse.sh

@@ -39,6 +39,17 @@ export FASTVIDEO_GENERATION_SEGMENT_CAP="${FASTVIDEO_GENERATION_SEGMENT_CAP:-6}"
 export FASTVIDEO_PROMPT_AUTO_SLEEP_MS="${FASTVIDEO_PROMPT_AUTO_SLEEP_MS:-120}"
 export FASTVIDEO_PROMPT_AUTO_TIMEOUT_MS="${FASTVIDEO_PROMPT_AUTO_TIMEOUT_MS:-1800}"
 
+if [[ "${ENABLE_TORCH_COMPILE}" == "1" ]]; then
+  # Persist Inductor, AOTAutograd, and Triton artifacts across launches.
+  export DREAMVERSE_TORCH_COMPILE_CACHE_ROOT="${DREAMVERSE_TORCH_COMPILE_CACHE_ROOT:-${HOME}/.cache/dreamverse/torch_compile}"
+  export TORCHINDUCTOR_CACHE_DIR="${TORCHINDUCTOR_CACHE_DIR:-${DREAMVERSE_TORCH_COMPILE_CACHE_ROOT}/inductor}"
+  export TRITON_CACHE_DIR="${TRITON_CACHE_DIR:-${DREAMVERSE_TORCH_COMPILE_CACHE_ROOT}/triton}"
+  export TORCHINDUCTOR_FX_GRAPH_CACHE="${TORCHINDUCTOR_FX_GRAPH_CACHE:-1}"
+  export TORCHINDUCTOR_AUTOGRAD_CACHE="${TORCHINDUCTOR_AUTOGRAD_CACHE:-1}"
+  mkdir -p "${TORCHINDUCTOR_CACHE_DIR}" "${TRITON_CACHE_DIR}"
+  echo "[launch-demo] torch.compile cache: ${DREAMVERSE_TORCH_COMPILE_CACHE_ROOT}"
+fi
+
 cd "${DREAMVERSE_ROOT}"
 
 if ! command -v dreamverse-server >/dev/null 2>&1; then

apps/dreamverse/scripts/modal/modal_app.py

@@ -8,12 +8,10 @@ import modal
 
 IMAGE = os.environ.get("DREAMVERSE_IMAGE")
 if not IMAGE:
-    raise RuntimeError(
-        "DREAMVERSE_IMAGE is required. Set it to a published SHA-specific Dreamverse image, "
-        "for example a dreamverse-backend-cuda13.0.0-sha-* tag or a "
-        "dreamverse-ui-cuda13.0.0-sha-* tag if serving the static UI. "
-        "CUDA 12 / cu126 images use the corresponding cuda12.6.3 tag."
-    )
+    raise RuntimeError("DREAMVERSE_IMAGE is required. Set it to a published SHA-specific Dreamverse image, "
+                       "for example a dreamverse-backend-cuda13.0.0-sha-* tag or a "
+                       "dreamverse-ui-cuda13.0.0-sha-* tag if serving the static UI. "
+                       "CUDA 12 / cu126 images use the corresponding cuda12.6.3 tag.")
 
 # ``@modal.web_server`` invokes ``serve()`` directly and bypasses the image
 # ENTRYPOINT (``docker/docker_entrypoint.sh``).  That entrypoint normally
@@ -65,14 +63,10 @@ def serve():
     # ``or ""`` collapses ``None`` (unset) into an empty string, ``.strip()``
     # collapses whitespace-only values (e.g. ``"   "``) — both should be
     # treated as missing.
-    missing = [
-        k for k in _REQUIRED_SECRET_KEYS
-        if not (os.environ.get(k) or "").strip()
-    ]
+    missing = [k for k in _REQUIRED_SECRET_KEYS if not (os.environ.get(k) or "").strip()]
     if missing:
-        raise RuntimeError(
-            "dreamverse-api-keys secret is missing required entries: "
-            f"{', '.join(missing)}. Add them with `modal secret create "
-            "dreamverse-api-keys ... --force` and redeploy "
-            "(see apps/dreamverse/scripts/modal/README.md).")
+        raise RuntimeError("dreamverse-api-keys secret is missing required entries: "
+                           f"{', '.join(missing)}. Add them with `modal secret create "
+                           "dreamverse-api-keys ... --force` and redeploy "
+                           "(see apps/dreamverse/scripts/modal/README.md).")
     subprocess.Popen(["dreamverse-server", "--host", "0.0.0.0", "--port", "8009"])

apps/fastvideo_studio/.env.example

@@ -1 +1 @@
-PUBLIC_API_BASE_URL=http://localhost:8189/api
+NEXT_PUBLIC_API_BASE_URL=http://localhost:8189/api

apps/fastvideo_studio/.gitignore

@@ -13,6 +13,12 @@
 # testing
 /coverage
 
+# playwright
+/test-results/
+/playwright-report/
+/blob-report/
+/.playwright/
+
 # next.js
 /.next/
 /out/
@@ -39,4 +45,3 @@ yarn-error.log*
 # typescript
 *.tsbuildinfo
 next-env.d.ts
-.svelte-kit

apps/fastvideo_studio/.prettierrc

@@ -1,4 +0,0 @@
-{
-	"useTabs": true,
-	"tabWidth": 4
-}

apps/fastvideo_studio/README.md

@@ -11,19 +11,63 @@ The UI currently supports:
 - Datasets
 - Gallery View
 
+## Project Structure
+
+```
+apps/fastvideo_studio/
+├── server.py / job_runner.py / database.py   # FastAPI backend + job lifecycle
+├── mock_server.py                            # In-memory API mock for e2e tests
+├── models/                                   # Pydantic request models (shared with the mock)
+├── training_config.py                        # Studio workloads → fastvideo/train YAML configs
+├── tests/                                    # Backend unit tests (pytest)
+├── e2e/                                      # Playwright specs (run against the mock)
+└── src/
+    ├── app/                                  # Next.js App Router pages (thin routes)
+    ├── components/
+    │   ├── shell/                            # App chrome: header, sidebars, layout
+    │   ├── jobs/                             # Job queue, cards, create-job modal, log sidebar
+    │   ├── datasets/                         # Dataset cards, upload, captions
+    │   └── ui/                               # shadcn-style primitives (shared theme)
+    ├── stores/                               # Framework-agnostic state + React bridge (hooks/)
+    ├── lib/                                  # API client, types, option persistence
+    └── test/                                 # Vitest setup + factories
+```
+
+The visual theme (slate light/dark palettes, IBM Plex type, `#356cff` accent)
+is shared with `apps/dreamverse`; the toggle in the header persists the choice
+per browser.
+
+## Testing
+
+```bash
+npm run typecheck        # tsc
+npm test                 # vitest unit tests
+npm run e2e              # Playwright against the in-memory mock backend
+python -m pytest tests/  # backend unit tests (from apps/fastvideo_studio)
+```
+
 ## Quick Start
 
-The easiest way to run the application is:
+For local development, install dependencies and start the Next.js dev server:
+
+```bash
+cd apps/fastvideo_studio
+npm install
+npm run dev
+```
+
+You can then access the app at [http://localhost:3000](http://localhost:3000).
+Start the Python API server (default port 8189) in a separate terminal — see below.
+
+For a production build of the web app together with the API server:
 
 ```bash
 cd apps/fastvideo_studio
 npm install
 npm run build
-npm run start
+npm run start:all
 ```
 
-You can then access the app at [http://localhost:3000](http://localhost:3000).
-
 ### Running API and Web Separately
 
 The UI is composed of two separate components:
@@ -31,7 +75,7 @@ The UI is composed of two separate components:
 - API Server
 - Web Server
 
-To run each component separately, you can use the commands `npm run start:web` and `npm run start:api`.
+To run each component separately, you can use the commands `npm run start:web` (after `npm run build`) and `npm run start:api`.
 
 You can also run the API server with this command from the `apps/` directory:
 

apps/fastvideo_studio/database.py

@@ -9,6 +9,7 @@ from __future__ import annotations
 
 import contextlib
 import logging
+import json
 import sqlite3
 import threading
 from pathlib import Path
@@ -46,8 +47,13 @@ DEFAULT_SETTINGS: dict[str, Any] = {
 
 
 def _sqlite_row_get(row: sqlite3.Row, key: str, default: Any) -> Any:
-    """Like dict.get for sqlite3.Row (Row has no .get on Python 3.10)."""
-    return row[key] if key in row else default  # noqa: SIM401
+    """Like dict.get for sqlite3.Row (Row has no .get on Python 3.10).
+
+    NOTE: `key in row` tests Row *values*, not column names, so the membership
+    check has to go through .keys() -- otherwise every lookup falls back to the
+    default and jobs restored from the database lose their stored fields.
+    """
+    return row[key] if key in row.keys() else default  # noqa: SIM401, SIM118
 
 
 def _get_db_path(data_dir: Path) -> Path:
@@ -83,6 +89,9 @@ def _migrate_db(conn: sqlite3.Connection) -> None:
     _add_column_if_missing(conn, "jobs", "fps", "INTEGER", "24")
     _add_column_if_missing(conn, "jobs", "workload_type", "TEXT", "'t2v'")
     _add_column_if_missing(conn, "jobs", "image_path", "TEXT", "''")
+    _add_column_if_missing(conn, "jobs", "name", "TEXT", "''")
+    _add_column_if_missing(conn, "jobs", "last_image_path", "TEXT", "''")
+    _add_column_if_missing(conn, "jobs", "references_json", "TEXT", "''")
     _add_column_if_missing(conn, "jobs", "job_type", "TEXT", "'inference'")
     _add_column_if_missing(conn, "jobs", "data_path", "TEXT", "''")
     _add_column_if_missing(conn, "jobs", "max_train_steps", "INTEGER", "1000")
@@ -242,7 +251,8 @@ class Database:
         self._execute(
             """
             INSERT INTO jobs (
-                id, model_id, prompt, workload_type, image_path, job_type, status,
+                id, model_id, name, prompt, workload_type, image_path,
+                last_image_path, references_json, job_type, status,
                 created_at, started_at, finished_at, error, output_path, log_file_path,
                 num_inference_steps, num_frames, height, width, guidance_scale,
                 guidance_rescale, fps, seed, num_gpus, dit_cpu_offload,
@@ -251,18 +261,20 @@ class Database:
                 sp_size, negative_prompt,
                 data_path, max_train_steps, train_batch_size, learning_rate,
                 num_latent_t, validation_dataset_file, lora_rank,
-                ltx2_first_frame_conditioning_p,
                 dmd_use_vsa, dmd_vsa_sparsity, dmd_denoising_steps,
-                min_timestep_ratio, max_timestep_ratio, real_score_guidance_scale,
+                real_score_guidance_scale,
                 generator_update_interval, real_score_model_path, fake_score_model_path
             ) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
             """,
             (
                 job["id"],
                 job["model_id"],
+                job.get("name", ""),
                 job["prompt"],
                 job.get("workload_type", "t2v"),
                 job.get("image_path", ""),
+                job.get("last_image_path", ""),
+                json.dumps(job.get("references") or []),
                 job.get("job_type", "inference"),
                 job["status"],
                 job["created_at"],
@@ -297,12 +309,9 @@ class Database:
                 job.get("num_latent_t", 20),
                 job.get("validation_dataset_file", ""),
                 job.get("lora_rank", 32),
-                job.get("ltx2_first_frame_conditioning_p"),
                 1 if job.get("dmd_use_vsa") else 0,
                 job.get("dmd_vsa_sparsity", 0.8),
                 job.get("dmd_denoising_steps", "1000,757,522"),
-                job.get("min_timestep_ratio", 0.02),
-                job.get("max_timestep_ratio", 0.98),
                 job.get("real_score_guidance_scale", 3.5),
                 job.get("generator_update_interval", 5),
                 job.get("real_score_model_path", ""),
@@ -401,24 +410,6 @@ class Database:
         for file_name, caption in captions.items():
             self.upsert_dataset_caption(dataset_id, file_name, caption)
 
-    def update_dataset(self, dataset_id: str, updates: dict[str, Any]) -> None:
-        """Update dataset fields. Only provided keys are updated."""
-        if not updates:
-            return
-        allowed: set[str] = set()
-        cols = []
-        vals = []
-        for k, v in updates.items():
-            if k in allowed:
-                cols.append(f"{k} = ?")
-                vals.append(v)
-        if not cols:
-            return
-        vals.append(dataset_id)
-        sql = f"UPDATE datasets SET {', '.join(cols)} WHERE id = ?"
-        self._execute(sql, tuple(vals))
-        self._commit()
-
     def delete_dataset(self, dataset_id: str) -> bool:
         """Delete a dataset. Returns True if a row was deleted."""
         cur = self._execute("DELETE FROM datasets WHERE id = ?", (dataset_id, ))
@@ -443,7 +434,7 @@ class Database:
         cur = self._execute("SELECT * FROM settings WHERE id = 1")
         row = cur.fetchone()
         if not row:
-            return _default_settings_dict()
+            return default_settings_dict()
         t2v = ((row["default_model_id_t2v"] or row["default_model_id"] or "") if "default_model_id_t2v" in row else
                (row["default_model_id"] or ""))
         i2v = ((row["default_model_id_i2v"] or "") if "default_model_id_i2v" in row else "")
@@ -557,16 +548,17 @@ def _row_to_job(row: sqlite3.Row) -> dict[str, Any]:
     }
     float_defaults = {
         "dmd_vsa_sparsity": 0.8,
-        "min_timestep_ratio": 0.02,
-        "max_timestep_ratio": 0.98,
         "real_score_guidance_scale": 3.5,
     }
     result = {
         "id": row["id"],
         "model_id": row["model_id"],
+        "name": _sqlite_row_get(row, "name", "") or "",
         "prompt": row["prompt"],
         "workload_type": _sqlite_row_get(row, "workload_type", "t2v"),
         "image_path": _sqlite_row_get(row, "image_path", "") or "",
+        "last_image_path": _sqlite_row_get(row, "last_image_path", "") or "",
+        "references": _sqlite_row_get(row, "references_json", "") or "",
         "job_type": _sqlite_row_get(row, "job_type", "inference"),
         "status": row["status"],
         "created_at": row["created_at"],
@@ -625,7 +617,7 @@ def _row_to_job(row: sqlite3.Row) -> dict[str, Any]:
     return result
 
 
-def _default_settings_dict() -> dict[str, Any]:
+def default_settings_dict() -> dict[str, Any]:
     """Return default settings as API-style dict."""
     return {
         "defaultModelId": DEFAULT_SETTINGS["default_model_id"],

apps/fastvideo_studio/e2e/create-job-interaction.spec.ts

@@ -0,0 +1,63 @@
+import { expect, test } from '@playwright/test';
+
+import { skipWithoutMock } from './helpers';
+
+test.describe('create job interactions', () => {
+  skipWithoutMock();
+
+  for (const jobType of ['inference', 'finetuning', 'distillation']) {
+    test(`${jobType} remains interactive after repeated dialog dismissals`, async ({ page }) => {
+      await page.goto(`/${jobType}`);
+      const trigger = page.getByRole('button', { name: 'Create Job', exact: true });
+      const dialog = page.getByRole('dialog');
+
+      // Exercise both dismissal paths and reopen without reloading the page.
+      for (const closeWithEscape of [false, true]) {
+        await trigger.click();
+        await page.getByRole('menuitem').first().click();
+        await expect(dialog).toBeVisible();
+        if (closeWithEscape) {
+          await page.keyboard.press('Escape');
+        } else {
+          await dialog.getByRole('button', { name: 'Close', exact: true }).click();
+        }
+        await expect(dialog).toBeHidden();
+        await expect(page.locator('body')).toHaveCSS('pointer-events', 'auto');
+        await expect(trigger).toBeFocused();
+      }
+
+      await page.getByRole('link', { name: 'Datasets', exact: true }).click();
+      await expect(page).toHaveURL(/\/datasets$/);
+    });
+  }
+
+  test('preserves keyboard menu dismissal and dialog focus trapping', async ({ page }) => {
+    await page.goto('/inference');
+    const trigger = page.getByRole('button', { name: 'Create Job', exact: true });
+    await trigger.focus();
+    await page.keyboard.press('Enter');
+    const firstItem = page.getByRole('menuitem').first();
+    await expect(firstItem).toBeFocused();
+    await page.keyboard.press('Escape');
+    await expect(page.getByRole('menu')).toBeHidden();
+    await expect(trigger).toBeFocused();
+    await expect(page.locator('body')).toHaveCSS('pointer-events', 'auto');
+
+    await page.keyboard.press('Enter');
+    await expect(firstItem).toBeFocused();
+    await page.keyboard.press('Enter');
+    const dialog = page.getByRole('dialog');
+    await expect(dialog).toBeVisible();
+    await expect(dialog.getByLabel('Name (optional)')).toBeFocused();
+
+    // Shift+Tab from the first field wraps to Close, then Tab wraps back.
+    await page.keyboard.press('Shift+Tab');
+    await expect(dialog.getByRole('button', { name: 'Close', exact: true })).toBeFocused();
+    await page.keyboard.press('Tab');
+    await expect(dialog.getByLabel('Name (optional)')).toBeFocused();
+    await page.keyboard.press('Escape');
+    await expect(dialog).toBeHidden();
+    await expect(trigger).toBeFocused();
+    await expect(page.locator('body')).toHaveCSS('pointer-events', 'auto');
+  });
+});

apps/fastvideo_studio/e2e/create-job.spec.ts

@@ -0,0 +1,58 @@
+import { expect, test } from '@playwright/test';
+
+import { API_BASE, skipWithoutMock } from './helpers';
+
+/**
+ * Create-job flow: open the Create Job modal on /inference, fill the prompt
+ * (the model auto-selects once the mock's /api/models loads), submit, and
+ * confirm the new job lands in the queue.
+ */
+test.describe('create inference job', () => {
+  skipWithoutMock();
+
+  test('creates a T2V job and starts it without refreshing', async ({ page, request }) => {
+    await request.put(`${API_BASE}/settings`, { data: { autoStartJob: false } });
+    await page.goto('/inference');
+
+    // The trigger opens a real menu on click, so this path works for touch,
+    // mouse, and keyboard users.
+    await page.getByRole('button', { name: /create job/i }).click();
+    const t2vItem = page.getByRole('menuitem', { name: /T2V/i });
+    await expect(t2vItem).toBeVisible();
+    await t2vItem.click();
+
+    const dialog = page.getByRole('dialog');
+    await expect(dialog).toBeVisible();
+
+    // Wait for the mock's model catalogue to populate the dropdown (more than
+    // just the disabled placeholder), then pick one explicitly — the app's
+    // auto-selection is racy.
+    const modelSelect = dialog.getByLabel('Model', { exact: true });
+    await expect(modelSelect.locator('option')).not.toHaveCount(1);
+    await modelSelect.selectOption({ index: 1 });
+
+    const prompt = `e2e raccoon in sunflowers ${Date.now()}`;
+    await dialog.getByLabel('Prompt', { exact: true }).fill(prompt);
+
+    await dialog.getByRole('button', { name: 'Create Job' }).click();
+
+    // Modal closes and the queue refreshes with the newly created job.
+    await expect(dialog).toBeHidden();
+    await expect(page.getByText(prompt)).toBeVisible();
+    await expect(page.locator('body')).toHaveCSS('pointer-events', 'auto');
+
+    const card = page.getByRole('article').filter({ hasText: prompt });
+    await expect(card.getByText('pending', { exact: true })).toBeVisible();
+    const started = page.waitForResponse((response) =>
+      response.url().startsWith(`${API_BASE}/jobs/`) &&
+      response.url().endsWith('/start') &&
+      response.request().method() === 'POST',
+    );
+    await card.getByRole('button', { name: 'Start', exact: true }).click();
+    expect((await started).ok()).toBe(true);
+    await expect(card.getByText('running', { exact: true })).toBeVisible();
+
+    await page.getByRole('link', { name: 'Datasets', exact: true }).click();
+    await expect(page).toHaveURL(/\/datasets$/);
+  });
+});

apps/fastvideo_studio/e2e/datasets.spec.ts

@@ -0,0 +1,37 @@
+import { expect, test } from '@playwright/test';
+
+import { API_BASE, skipWithoutMock } from './helpers';
+
+/**
+ * Datasets page: the seeded datasets render as cards, and the Create Dataset
+ * modal opens from the header action.
+ */
+test.describe('datasets', () => {
+  skipWithoutMock();
+
+  test('lists the seeded datasets', async ({ page, request }) => {
+    // Read the seeded names from the mock so the assertion tracks the fixture.
+    const res = await request.get(`${API_BASE}/datasets`);
+    const datasets = (await res.json()) as Array<{ name: string }>;
+    expect(datasets.length).toBeGreaterThan(0);
+
+    await page.goto('/datasets');
+
+    for (const ds of datasets) {
+      await expect(page.getByText(ds.name, { exact: true })).toBeVisible();
+    }
+  });
+
+  test('opens the Create Dataset modal', async ({ page }) => {
+    await page.goto('/datasets');
+
+    await page.getByRole('button', { name: 'Add Dataset' }).click();
+
+    const dialog = page.getByRole('dialog');
+    await expect(dialog).toBeVisible();
+    await expect(
+      dialog.getByRole('heading', { name: /Add Dataset/i }),
+    ).toBeVisible();
+    await expect(dialog.getByLabel('Name', { exact: true })).toBeVisible();
+  });
+});

apps/fastvideo_studio/e2e/gallery.spec.ts

@@ -0,0 +1,44 @@
+import { expect, test } from '@playwright/test';
+
+import { API_BASE, skipWithoutMock } from './helpers';
+
+/**
+ * Gallery page: the seeded completed inference job surfaces as a media tile
+ * with playback controls or an explicit media-error fallback.
+ */
+test.describe('gallery', () => {
+  skipWithoutMock();
+
+  test('shows a media tile for the seeded completed job', async ({
+    page,
+    request,
+  }) => {
+    const res = await request.get(`${API_BASE}/jobs?job_type=inference`);
+    const jobs = (await res.json()) as Array<{
+      status: string;
+      output_path: string | null;
+      prompt: string;
+    }>;
+    const completed = jobs.find(
+      (j) => j.status === 'completed' && j.output_path,
+    );
+    expect(completed, 'mock should seed a completed inference job').toBeTruthy();
+
+    await page.goto('/gallery');
+
+    await expect(
+      page.getByRole('heading', { level: 1, name: 'Gallery' }),
+    ).toBeVisible();
+
+    const tile = page.locator('article').filter({ hasText: completed!.prompt });
+    await expect(tile).toBeVisible();
+    await expect(
+      tile.locator('video').or(tile.getByText('Preview unavailable')),
+    ).toBeVisible();
+
+    const video = tile.locator('video');
+    if (await video.isVisible()) {
+      await expect(video).toHaveAttribute('controls', '');
+    }
+  });
+});

apps/fastvideo_studio/e2e/gpus.spec.ts

@@ -0,0 +1,30 @@
+import { expect, test } from '@playwright/test';
+
+import { skipWithoutMock } from './helpers';
+
+/**
+ * GPU status page: the mock's fake GPUs render as cards with meters, and the
+ * page is reachable from the primary sidebar.
+ */
+test.describe('gpus', () => {
+  skipWithoutMock();
+
+  test('lists the mock GPUs with utilization meters', async ({ page }) => {
+    await page.goto('/gpus');
+
+    await expect(
+      page.getByRole('heading', { level: 1, name: 'GPUs' }),
+    ).toBeVisible();
+
+    await expect(page.getByText('GPU 0')).toBeVisible();
+    await expect(page.getByText('GPU 1')).toBeVisible();
+    expect(await page.getByRole('meter').count()).toBe(4);
+  });
+
+  test('is reachable from the sidebar', async ({ page }) => {
+    await page.goto('/inference');
+    await page.getByRole('link', { name: 'GPUs' }).click();
+    await expect(page).toHaveURL(/\/gpus$/);
+    await expect(page.getByText('GPU 0')).toBeVisible();
+  });
+});

apps/fastvideo_studio/e2e/helpers.ts

@@ -0,0 +1,44 @@
+import { test, type APIRequestContext } from '@playwright/test';
+
+/**
+ * Port and base URL of the mock API — the single source of truth shared by
+ * playwright.config.ts (which boots the mock and points `npm run dev` at it
+ * via NEXT_PUBLIC_API_BASE_URL) and the specs (which probe/read mock state
+ * directly through the Playwright `request` fixture).
+ *
+ * The default port is deliberately NOT the real server's default (8189): the
+ * e2e specs create and (if autostart is on) launch jobs, so running them
+ * against a real backend would mutate the developer's database and fire real
+ * GPU runs. API_BASE is derived solely from MOCK_API_PORT (it does not inherit
+ * NEXT_PUBLIC_API_BASE_URL) so the specs always talk to the mock.
+ */
+export const MOCK_API_PORT = process.env.MOCK_API_PORT || '8190';
+
+export const API_BASE = `http://127.0.0.1:${MOCK_API_PORT}/api`;
+
+export const MOCK_SKIP_MESSAGE =
+  `Mock backend not reachable at ${API_BASE}/__mock__ ` +
+  `(start fastvideo_studio.mock_server on port ${MOCK_API_PORT}).`;
+
+/**
+ * True only when the mock server answers on the configured port. Probes the
+ * mock-only /api/__mock__ sentinel (which the real server does not serve) so
+ * the suite refuses to run its mutating specs against a real backend.
+ */
+export async function mockIsUp(request: APIRequestContext): Promise<boolean> {
+  try {
+    const res = await request.get(`${API_BASE}/__mock__`);
+    if (!res.ok()) return false;
+    const body = await res.json();
+    return body?.mock === true;
+  } catch {
+    return false;
+  }
+}
+
+/** Self-skip every test in the enclosing describe when the mock is down. */
+export function skipWithoutMock(): void {
+  test.beforeEach(async ({ request }) => {
+    test.skip(!(await mockIsUp(request)), MOCK_SKIP_MESSAGE);
+  });
+}

apps/fastvideo_studio/e2e/shell.spec.ts

@@ -0,0 +1,115 @@
+import { expect, test } from '@playwright/test';
+
+import { skipWithoutMock } from './helpers';
+
+/**
+ * App-shell smoke: the Next.js frontend hydrates, renders the FastVideo logo
+ * and the primary-sidebar navigation, and routing between the top-level
+ * sections works. Each spec self-skips when the mock backend isn't reachable.
+ */
+test.describe('app shell', () => {
+  skipWithoutMock();
+
+  test('loads with the logo and primary-sidebar nav', async ({ page }) => {
+    await page.goto('/');
+    // The root route redirects to /inference.
+    await expect(page).toHaveURL(/\/inference$/);
+
+    await expect(page.getByRole('img', { name: /fastvideo/i })).toBeVisible();
+
+    for (const label of ['Inference', 'Datasets', 'Gallery', 'Settings']) {
+      await expect(page.getByRole('link', { name: label })).toBeVisible();
+    }
+  });
+
+  test('navigates between the primary sections', async ({ page }) => {
+    await page.goto('/inference');
+    await expect(
+      page.getByRole('heading', { level: 1, name: 'Jobs' }),
+    ).toBeVisible();
+
+    const sections: Array<{ link: string; url: RegExp; title: string }> = [
+      { link: 'Datasets', url: /\/datasets$/, title: 'Datasets' },
+      { link: 'Gallery', url: /\/gallery$/, title: 'Gallery' },
+      { link: 'Settings', url: /\/settings$/, title: 'Settings' },
+      { link: 'Inference', url: /\/inference$/, title: 'Jobs' },
+    ];
+
+    for (const section of sections) {
+      await page.getByRole('link', { name: section.link }).click();
+      await expect(page).toHaveURL(section.url);
+      // The header <h1> is the only level-1 heading and reflects the route.
+      await expect(
+        page.getByRole('heading', { level: 1, name: section.title }),
+      ).toBeVisible();
+      await expect(page.getByRole('main')).toHaveCount(1);
+    }
+  });
+
+  test('keeps navigation and content usable at responsive breakpoints', async ({
+    page,
+  }) => {
+    for (const width of [320, 375, 414, 768]) {
+      await page.setViewportSize({ width, height: 800 });
+      await page.goto('/inference');
+
+      const main = page.getByRole('main');
+      await expect(main).toBeVisible();
+      await expect(
+        page.getByRole('button', { name: /Create Job/i }),
+      ).toBeVisible();
+
+      const initialBox = await main.boundingBox();
+      expect(initialBox?.x).toBe(width < 768 ? 0 : 220);
+      expect(initialBox?.width).toBe(width < 768 ? width : width - 220);
+
+      const navigation = page.getByRole('navigation', {
+        name: 'Primary navigation',
+      });
+      if (width < 768) {
+        await expect(
+          page.getByRole('button', { name: 'Open navigation' }),
+        ).toBeVisible();
+        await page.getByRole('button', { name: 'Open navigation' }).click();
+      }
+      await expect(navigation).toBeVisible();
+      await navigation.getByRole('link', { name: 'Datasets' }).click();
+
+      await expect(page).toHaveURL(/\/datasets$/);
+      expect(
+        await page.evaluate(
+          () => document.documentElement.scrollWidth <= window.innerWidth,
+        ),
+      ).toBe(true);
+    }
+  });
+
+  test('uses full-width detail drawers on mobile', async ({ page }) => {
+    await page.setViewportSize({ width: 320, height: 800 });
+    await page.goto('/inference');
+
+    await page
+      .locator('article button[aria-pressed="false"]')
+      .first()
+      .click();
+    const jobDrawer = page.getByRole('dialog', { name: 'Job details' });
+    await expect(jobDrawer).toBeVisible();
+    expect(await jobDrawer.boundingBox()).toMatchObject({ x: 0, width: 320 });
+    await jobDrawer.getByRole('button', { name: 'Close' }).click();
+
+    await page.goto('/datasets');
+    await page
+      .locator('article button[aria-pressed="false"]')
+      .first()
+      .click();
+
+    const datasetDrawer = page.getByRole('dialog', {
+      name: /dataset details$/,
+    });
+    await expect(datasetDrawer).toBeVisible();
+    expect(await datasetDrawer.boundingBox()).toMatchObject({
+      x: 0,
+      width: 320,
+    });
+  });
+});

apps/fastvideo_studio/eslint.config.mjs

@@ -1,2 +0,0 @@
-// Minimal ESLint config for SvelteKit (add typescript-eslint / eslint-plugin-svelte as needed)
-export default [];

apps/fastvideo_studio/gpu.py

@@ -0,0 +1,74 @@
+# SPDX-License-Identifier: Apache-2.0
+"""GPU telemetry for the studio status page, via NVML (nvidia-ml-py)."""
+
+from __future__ import annotations
+
+import contextlib
+import logging
+from typing import Any
+
+logger = logging.getLogger("fastvideo.studio.gpu")
+
+_nvml_initialized = False
+
+
+def _ensure_nvml() -> Any:
+    """Import and initialize NVML once; raises on machines without it."""
+    global _nvml_initialized  # noqa: PLW0603
+    import pynvml
+    if not _nvml_initialized:
+        pynvml.nvmlInit()
+        _nvml_initialized = True
+    return pynvml
+
+
+def _device_snapshot(pynvml: Any, index: int) -> dict[str, Any]:
+    handle = pynvml.nvmlDeviceGetHandleByIndex(index)
+    name = pynvml.nvmlDeviceGetName(handle)
+    if isinstance(name, bytes):
+        name = name.decode()
+    mem = pynvml.nvmlDeviceGetMemoryInfo(handle)
+    util = pynvml.nvmlDeviceGetUtilizationRates(handle)
+
+    # Optional sensors: not every GPU/driver exposes them.
+    temperature: int | None = None
+    power_watts: float | None = None
+    power_limit_watts: float | None = None
+    with contextlib.suppress(pynvml.NVMLError):
+        temperature = int(pynvml.nvmlDeviceGetTemperature(handle, pynvml.NVML_TEMPERATURE_GPU))
+    with contextlib.suppress(pynvml.NVMLError):
+        power_watts = pynvml.nvmlDeviceGetPowerUsage(handle) / 1000.0
+        power_limit_watts = (pynvml.nvmlDeviceGetEnforcedPowerLimit(handle) / 1000.0)
+
+    return {
+        "index": index,
+        "name": name,
+        "utilization": int(util.gpu),
+        "memory_used_mib": int(mem.used / (1024 * 1024)),
+        "memory_total_mib": int(mem.total / (1024 * 1024)),
+        "temperature_c": temperature,
+        "power_watts": power_watts,
+        "power_limit_watts": power_limit_watts,
+    }
+
+
+def get_gpu_snapshot() -> dict[str, Any]:
+    """Return {"available", "gpus", "error"} — never raises.
+
+    ``available: False`` covers both "no NVIDIA driver/library on this host"
+    and transient NVML failures; the frontend shows ``error`` as-is.
+    """
+    try:
+        pynvml = _ensure_nvml()
+        count = pynvml.nvmlDeviceGetCount()
+        gpus = [_device_snapshot(pynvml, i) for i in range(count)]
+        return {"available": True, "gpus": gpus, "error": None}
+    except ImportError:
+        return {
+            "available": False,
+            "gpus": [],
+            "error": "nvidia-ml-py is not installed on the API server host.",
+        }
+    except Exception as exc:  # NVMLError, driver issues, …
+        logger.warning("GPU snapshot failed: %s", exc)
+        return {"available": False, "gpus": [], "error": str(exc)}

apps/fastvideo_studio/job_runner.py

@@ -10,7 +10,9 @@ from __future__ import annotations
 import atexit
 import collections
 import contextlib
+import copy
 import enum
+import json
 import logging
 import logging.handlers
 import multiprocessing as mp
@@ -23,12 +25,13 @@ import time
 from dataclasses import dataclass, field
 from pathlib import Path
 from typing import Any
+import yaml
+
 from fastvideo.utils import get_mp_context
 from fastvideo_studio.database import Database
 from fastvideo_studio.training_config import (
-    build_training_args,
+    build_training_config,
     get_training_env,
-    get_training_module_info,
 )
 
 logger = logging.getLogger("fastvideo.studio.job_runner")
@@ -122,10 +125,13 @@ class LogBufferHandler(logging.Handler):
 class Job:
     id: str
     model_id: str
-    prompt: str
+    name: str = ""
+    prompt: str = ""
     workload_type: str = "t2v"
     job_type: str = "inference"
     image_path: str = ""
+    last_image_path: str = ""
+    references: list[dict[str, Any]] = field(default_factory=list)
     status: JobStatus = JobStatus.PENDING
     created_at: float = field(default_factory=time.time)
     started_at: float | None = None
@@ -144,6 +150,7 @@ class Job:
     negative_prompt: str = ""
     num_gpus: int = 1
     dit_cpu_offload: bool = False
+    dit_layerwise_offload: bool = False
     text_encoder_cpu_offload: bool = False
     vae_cpu_offload: bool = False
     image_encoder_cpu_offload: bool = False
@@ -160,13 +167,10 @@ class Job:
     num_latent_t: int = 20
     validation_dataset_file: str = ""
     lora_rank: int = 32
-    ltx2_first_frame_conditioning_p: float | None = None
     # DMD options
     dmd_use_vsa: bool = False
     dmd_vsa_sparsity: float = 0.8
     dmd_denoising_steps: str = "1000,757,522"
-    min_timestep_ratio: float = 0.02
-    max_timestep_ratio: float = 0.98
     real_score_guidance_scale: float = 3.5
     generator_update_interval: int = 5
     real_score_model_path: str = ""
@@ -182,10 +186,13 @@ class Job:
         return {
             "id": self.id,
             "model_id": self.model_id,
+            "name": self.name,
             "prompt": self.prompt,
             "workload_type": self.workload_type,
             "job_type": self.job_type,
             "image_path": self.image_path,
+            "last_image_path": self.last_image_path,
+            "references": self.references,
             "status": self.status.value,
             "created_at": self.created_at,
             "started_at": self.started_at,
@@ -204,6 +211,7 @@ class Job:
             "negative_prompt": self.negative_prompt,
             "num_gpus": self.num_gpus,
             "dit_cpu_offload": self.dit_cpu_offload,
+            "dit_layerwise_offload": self.dit_layerwise_offload,
             "text_encoder_cpu_offload": self.text_encoder_cpu_offload,
             "vae_cpu_offload": self.vae_cpu_offload,
             "image_encoder_cpu_offload": self.image_encoder_cpu_offload,
@@ -221,12 +229,9 @@ class Job:
             "num_width": self.width,
             "validation_dataset_file": self.validation_dataset_file,
             "lora_rank": self.lora_rank,
-            "ltx2_first_frame_conditioning_p": self.ltx2_first_frame_conditioning_p,
             "dmd_use_vsa": self.dmd_use_vsa,
             "dmd_vsa_sparsity": self.dmd_vsa_sparsity,
             "dmd_denoising_steps": self.dmd_denoising_steps,
-            "min_timestep_ratio": self.min_timestep_ratio,
-            "max_timestep_ratio": self.max_timestep_ratio,
             "real_score_guidance_scale": self.real_score_guidance_scale,
             "generator_update_interval": self.generator_update_interval,
             "real_score_model_path": self.real_score_model_path or "",
@@ -237,6 +242,75 @@ class Job:
         }
 
 
+MINIMAX_H3_REF2VA_PIPELINE = "MiniMaxH3Ref2VAModularPipeline"
+
+
+def _build_h3_references(raw: list[dict[str, Any]]) -> list[Any]:
+    """Turn the API's reference dicts into MiniMaxH3Reference objects.
+
+    Imported lazily so the API server starts without pulling in fastvideo.
+    """
+    from fastvideo.pipelines.basic.minimax_h3 import MiniMaxH3Reference
+
+    built = []
+    for i, ref in enumerate(raw):
+        source = (ref or {}).get("source")
+        if not source:
+            raise ValueError(f"reference {i} has no source")
+        if not os.path.isfile(source):
+            raise ValueError(f"reference {i} source not found: {source}")
+        kwargs: dict[str, Any] = {
+            "source": source,
+            "media_type": (ref.get("media_type") or "image"),
+        }
+        for opt in ("soundtrack", "fps", "sample_rate"):
+            if ref.get(opt) not in (None, ""):
+                kwargs[opt] = ref[opt]
+        built.append(MiniMaxH3Reference(**kwargs))
+    return built
+
+
+JOB_LOG_FILENAME = "out.log"
+
+
+def _job_log_path(output_dir: str, job_id: str) -> str:
+    """Each job's log lives beside its outputs: <output_dir>/<job_id>/out.log."""
+    return os.path.join(output_dir, job_id, JOB_LOG_FILENAME)
+
+
+def _decode_references(value: Any) -> list[dict[str, Any]]:
+    """Reference lists round-trip through the DB as JSON text."""
+    if not value:
+        return []
+    if isinstance(value, list):
+        return list(value)
+    try:
+        decoded = json.loads(value)
+    except (TypeError, ValueError):
+        logger.warning("Could not decode stored references: %r", value)
+        return []
+    return list(decoded) if isinstance(decoded, list) else []
+
+
+def _generator_is_alive(generator: Any) -> bool:
+    """True if the generator's worker processes are all still running.
+
+    A cached VideoGenerator holds a MultiprocExecutor whose workers are separate
+    processes; nothing notices when they exit. Probing `proc.is_alive()` is what
+    the executor itself uses during shutdown. Anything unexpected in the object
+    graph is treated as alive so a probe failure can never wedge the cache.
+    """
+    executor = getattr(generator, "executor", None)
+    workers = getattr(executor, "workers", None)
+    if not workers:
+        return True
+    try:
+        return all(w.proc.is_alive() for w in workers)
+    except Exception:
+        logger.debug("Worker liveness probe failed", exc_info=True)
+        return True
+
+
 class JobRunner:
     """Manages video generation jobs, their execution, and generator caching."""
 
@@ -281,7 +355,7 @@ class JobRunner:
         """Populate job's log buffer from its log file if it exists."""
         path = job.log_file_path
         if not path:
-            path = os.path.join(self.log_dir, f"{job.id}.log")
+            path = _job_log_path(self.output_dir, job.id)
         if not os.path.isfile(path):
             return
         try:
@@ -319,10 +393,13 @@ class JobRunner:
                 job = Job(
                     id=row["id"],
                     model_id=row["model_id"],
+                    name=row.get("name", "") or "",
                     prompt=row["prompt"],
                     workload_type=row.get("workload_type", "t2v"),
                     job_type=row.get("job_type", "inference"),
                     image_path=row.get("image_path", "") or "",
+                    last_image_path=row.get("last_image_path", "") or "",
+                    references=_decode_references(row.get("references")),
                     data_path=row.get("data_path", "") or "",
                     max_train_steps=row.get("max_train_steps", 1000),
                     train_batch_size=row.get("train_batch_size", 1),
@@ -330,12 +407,9 @@ class JobRunner:
                     num_latent_t=row.get("num_latent_t", 20),
                     validation_dataset_file=row.get("validation_dataset_file", "") or "",
                     lora_rank=row.get("lora_rank", 32),
-                    ltx2_first_frame_conditioning_p=row.get("ltx2_first_frame_conditioning_p"),
                     dmd_use_vsa=row.get("dmd_use_vsa", False),
                     dmd_vsa_sparsity=float(row.get("dmd_vsa_sparsity", 0.8)),
                     dmd_denoising_steps=row.get("dmd_denoising_steps", "1000,757,522") or "1000,757,522",
-                    min_timestep_ratio=float(row.get("min_timestep_ratio", 0.02)),
-                    max_timestep_ratio=float(row.get("max_timestep_ratio", 0.98)),
                     real_score_guidance_scale=float(row.get("real_score_guidance_scale", 3.5)),
                     generator_update_interval=int(row.get("generator_update_interval", 5)),
                     real_score_model_path=row.get("real_score_model_path", "") or "",
@@ -358,6 +432,7 @@ class JobRunner:
                     negative_prompt=row.get("negative_prompt", "") or "",
                     num_gpus=row.get("num_gpus", 1),
                     dit_cpu_offload=row.get("dit_cpu_offload", False),
+                    dit_layerwise_offload=row.get("dit_layerwise_offload", False),
                     text_encoder_cpu_offload=row.get("text_encoder_cpu_offload", False),
                     vae_cpu_offload=row.get("vae_cpu_offload", False),
                     image_encoder_cpu_offload=row.get("image_encoder_cpu_offload", False),
@@ -402,9 +477,12 @@ class JobRunner:
         job_id: str,
         model_id: str,
         prompt: str,
+        name: str = "",
         workload_type: str = "t2v",
         job_type: str = "inference",
         image_path: str = "",
+        last_image_path: str = "",
+        references: list[dict[str, Any]] | None = None,
         data_path: str = "",
         max_train_steps: int = 1000,
         train_batch_size: int = 1,
@@ -412,12 +490,9 @@ class JobRunner:
         num_latent_t: int = 20,
         validation_dataset_file: str = "",
         lora_rank: int = 32,
-        ltx2_first_frame_conditioning_p: float | None = None,
         dmd_use_vsa: bool = False,
         dmd_vsa_sparsity: float = 0.8,
         dmd_denoising_steps: str = "1000,757,522",
-        min_timestep_ratio: float = 0.02,
-        max_timestep_ratio: float = 0.98,
         real_score_guidance_scale: float = 3.5,
         generator_update_interval: int = 5,
         real_score_model_path: str = "",
@@ -433,6 +508,7 @@ class JobRunner:
         num_gpus: int = 1,
         negative_prompt: str = "",
         dit_cpu_offload: bool = False,
+        dit_layerwise_offload: bool = False,
         text_encoder_cpu_offload: bool = False,
         vae_cpu_offload: bool = False,
         image_encoder_cpu_offload: bool = False,
@@ -446,10 +522,13 @@ class JobRunner:
         job = Job(
             id=job_id,
             model_id=model_id,
+            name=(name or "").strip(),
             prompt=prompt.strip(),
             workload_type=workload_type or "t2v",
             job_type=job_type or "inference",
             image_path=image_path or "",
+            last_image_path=last_image_path or "",
+            references=list(references or []),
             data_path=data_path or "",
             max_train_steps=max_train_steps,
             train_batch_size=train_batch_size,
@@ -457,12 +536,9 @@ class JobRunner:
             num_latent_t=num_latent_t,
             validation_dataset_file=validation_dataset_file or "",
             lora_rank=lora_rank,
-            ltx2_first_frame_conditioning_p=ltx2_first_frame_conditioning_p,
             dmd_use_vsa=dmd_use_vsa,
             dmd_vsa_sparsity=dmd_vsa_sparsity,
             dmd_denoising_steps=dmd_denoising_steps,
-            min_timestep_ratio=min_timestep_ratio,
-            max_timestep_ratio=max_timestep_ratio,
             real_score_guidance_scale=real_score_guidance_scale,
             generator_update_interval=generator_update_interval,
             real_score_model_path=real_score_model_path or "",
@@ -478,6 +554,7 @@ class JobRunner:
             negative_prompt=negative_prompt or "",
             num_gpus=num_gpus,
             dit_cpu_offload=dit_cpu_offload,
+            dit_layerwise_offload=dit_layerwise_offload,
             text_encoder_cpu_offload=text_encoder_cpu_offload,
             vae_cpu_offload=vae_cpu_offload,
             image_encoder_cpu_offload=image_encoder_cpu_offload,
@@ -535,6 +612,84 @@ class JobRunner:
         logger.info("Deleted job %s", job.id)
         return True
 
+    CONFIG_FIELDS: tuple[str, ...] = (
+        "model_id",
+        "name",
+        "prompt",
+        "workload_type",
+        "job_type",
+        "image_path",
+        "last_image_path",
+        "references",
+        "negative_prompt",
+        "num_inference_steps",
+        "num_frames",
+        "height",
+        "width",
+        "guidance_scale",
+        "guidance_rescale",
+        "fps",
+        "seed",
+        "num_gpus",
+        "dit_cpu_offload",
+        "dit_layerwise_offload",
+        "text_encoder_cpu_offload",
+        "vae_cpu_offload",
+        "image_encoder_cpu_offload",
+        "use_fsdp_inference",
+        "enable_torch_compile",
+        "vsa_sparsity",
+        "tp_size",
+        "sp_size",
+        "data_path",
+        "max_train_steps",
+        "train_batch_size",
+        "learning_rate",
+        "num_latent_t",
+        "validation_dataset_file",
+        "lora_rank",
+        "dmd_use_vsa",
+        "dmd_vsa_sparsity",
+        "dmd_denoising_steps",
+        "real_score_guidance_scale",
+        "generator_update_interval",
+        "real_score_model_path",
+        "fake_score_model_path",
+    )
+
+    def duplicate_job(self, job_id: str, new_job_id: str) -> Job:
+        """Create a new pending job with an existing job's configuration.
+
+        Runtime state (status, timings, logs, outputs) is not carried over.
+        """
+        with self._jobs_lock:
+            source = self._jobs.get(job_id)
+        if source is None:
+            raise ValueError(f"Job {job_id} not found")
+        config = {f: copy.deepcopy(getattr(source, f)) for f in self.CONFIG_FIELDS}
+        return self.create_job(job_id=new_job_id, **config)
+
+    #: Editable exactly when startable: the same set start_job() accepts.
+    EDITABLE_STATUSES = (JobStatus.PENDING, JobStatus.FAILED, JobStatus.STOPPED)
+
+    def update_job_config(self, job_id: str, updates: dict[str, Any]) -> Job:
+        """Edit the configuration of a job that has not produced a result."""
+        with self._jobs_lock:
+            job = self._jobs.get(job_id)
+        if job is None:
+            raise ValueError(f"Job {job_id} not found")
+        if job.status not in self.EDITABLE_STATUSES:
+            allowed = ", ".join(s.value for s in self.EDITABLE_STATUSES)
+            raise ValueError(f"Job is {job.status.value}; only {allowed} jobs can be edited. "
+                             "Duplicate it instead.")
+        unknown = set(updates) - set(self.CONFIG_FIELDS)
+        if unknown:
+            raise ValueError(f"Not editable: {', '.join(sorted(unknown))}")
+        for field_name, value in updates.items():
+            setattr(job, field_name, value)
+        self._save_job(job)
+        return job
+
     def start_job(self, job_id: str) -> Job:
         """Start (or restart) a pending / stopped / failed job.
         
@@ -637,6 +792,8 @@ class JobRunner:
         workload_type: str,
         num_gpus: int,
         dit_cpu_offload: bool = False,
+        dit_layerwise_offload: bool = False,
+        override_pipeline_cls_name: str | None = None,
         text_encoder_cpu_offload: bool = False,
         vae_cpu_offload: bool = False,
         image_encoder_cpu_offload: bool = False,
@@ -652,6 +809,10 @@ class JobRunner:
             workload_type,
             num_gpus,
             dit_cpu_offload,
+            dit_layerwise_offload,
+            # Ref2VA loads different DiT weights (transformer_ref), so the
+            # override must key the cache or a t2v/i2v generator gets reused.
+            override_pipeline_cls_name,
             text_encoder_cpu_offload,
             vae_cpu_offload,
             image_encoder_cpu_offload,
@@ -664,8 +825,21 @@ class JobRunner:
 
         # Generators are cached by model_id and configuration parameters
         with self._generators_lock:
-            if cache_key in self._generators:
-                return self._generators[cache_key]
+            cached = self._generators.get(cache_key)
+            if cached is not None:
+                if _generator_is_alive(cached):
+                    return cached
+                # Workers can exit while a generator sits idle in the cache;
+                # reusing it fails every later job with the same config.
+                logger.warning(
+                    "Cached generator for %s has dead workers; reloading.",
+                    model_id,
+                )
+                self._generators.pop(cache_key, None)
+                try:
+                    cached.shutdown()
+                except Exception:
+                    logger.debug("Shutdown of the dead generator failed", exc_info=True)
 
         # Import lazily so starting the server is fast even without a GPU.
         from fastvideo import VideoGenerator
@@ -691,6 +865,11 @@ class JobRunner:
         gen = VideoGenerator.from_pretrained(
             model_id,
             workload_type=workload_type,
+            num_gpus=num_gpus,
+            dit_layerwise_offload=dit_layerwise_offload,
+            **({
+                "override_pipeline_cls_name": override_pipeline_cls_name
+            } if override_pipeline_cls_name else {}),
             dit_cpu_offload=dit_cpu_offload,
             text_encoder_cpu_offload=text_encoder_cpu_offload,
             vae_cpu_offload=vae_cpu_offload,
@@ -720,10 +899,9 @@ class JobRunner:
     def _run_training_job(self, job: Job):
         """Run a finetuning, distillation, or LoRA job via subprocess."""
         buf = job._log_buf
-        os.makedirs(self.log_dir, exist_ok=True)
-        job.log_file_path = os.path.join(self.log_dir, f"{job.id}.log")
         job_output_dir = os.path.join(self.output_dir, job.id)
         os.makedirs(job_output_dir, exist_ok=True)
+        job.log_file_path = _job_log_path(self.output_dir, job.id)
 
         if not job.data_path or not os.path.isdir(job.data_path):
             job.status = JobStatus.FAILED
@@ -733,46 +911,23 @@ class JobRunner:
             self._save_job(job)
             return
 
-        module_info = get_training_module_info(job.workload_type, job.model_id)
-        if not module_info:
+        env = os.environ.copy()
+        env.update(get_training_env())
+
+        try:
+            # Job.to_dict() carries every key the config builder reads
+            # (extra keys are ignored by its .get() lookups).
+            train_config = build_training_config(job.to_dict(), job_output_dir)
+        except ValueError as exc:
             job.status = JobStatus.FAILED
-            job.error = f"Unknown workload type: {job.workload_type}"
+            job.error = str(exc)
             job.finished_at = time.time()
             self._save_job(job)
             return
 
-        module_path, _pipeline_workload, use_vsa, _is_lora = module_info
-        dmd_use_vsa = (job.workload_type.startswith("dmd_") and getattr(job, "dmd_use_vsa", False))
-        env = os.environ.copy()
-        env.update(get_training_env(use_vsa or dmd_use_vsa))
-
-        job_dict = {
-            "model_id": job.model_id,
-            "data_path": job.data_path,
-            "workload_type": job.workload_type,
-            "num_gpus": job.num_gpus,
-            "max_train_steps": job.max_train_steps,
-            "train_batch_size": job.train_batch_size,
-            "learning_rate": job.learning_rate,
-            "num_latent_t": job.num_latent_t,
-            "num_height": job.height,
-            "num_width": job.width,
-            "num_frames": job.num_frames,
-            "validation_dataset_file": job.validation_dataset_file,
-            "lora_rank": job.lora_rank,
-            "ltx2_first_frame_conditioning_p": job.ltx2_first_frame_conditioning_p,
-        }
-        if job.workload_type.startswith("dmd_") or job.workload_type.startswith("self_forcing_"):
-            job_dict["dmd_use_vsa"] = getattr(job, "dmd_use_vsa", False)
-            job_dict["dmd_vsa_sparsity"] = getattr(job, "dmd_vsa_sparsity", 0.8)
-            job_dict["dmd_denoising_steps"] = getattr(job, "dmd_denoising_steps", "1000,757,522")
-            job_dict["min_timestep_ratio"] = getattr(job, "min_timestep_ratio", 0.02)
-            job_dict["max_timestep_ratio"] = getattr(job, "max_timestep_ratio", 0.98)
-            job_dict["real_score_guidance_scale"] = getattr(job, "real_score_guidance_scale", 3.5)
-            job_dict["generator_update_interval"] = getattr(job, "generator_update_interval", 5)
-            job_dict["real_score_model_path"] = (getattr(job, "real_score_model_path", "") or job.model_id)
-            job_dict["fake_score_model_path"] = (getattr(job, "fake_score_model_path", "") or job.model_id)
-        train_args = build_training_args(job_dict, job_output_dir)
+        config_path = os.path.join(job_output_dir, "train_config.yaml")
+        with open(config_path, "w", encoding="utf-8") as f:
+            yaml.safe_dump(train_config, f, sort_keys=False)
 
         repo_root = Path(__file__).resolve().parent.parent
         torchrun_cmd = [
@@ -783,9 +938,12 @@ class JobRunner:
             str(job.num_gpus),
             "--nnodes",
             "1",
-            str(repo_root / module_path),
-        ] + train_args
-        buf.write(f"Starting training: {' '.join(torchrun_cmd[:12])}...")
+            "-m",
+            "fastvideo.train.entrypoint.train",
+            "--config",
+            config_path,
+        ]
+        buf.write(f"Starting training: {' '.join(torchrun_cmd)}")
         buf.phase = "starting"
 
         try:
@@ -825,12 +983,21 @@ class JobRunner:
                 job._process.wait()
                 exit_code = job._process.returncode or 0
 
-            if exit_code == 0:
+            if job._stop_event.is_set():
+                # Terminated by stop_job() without the reader loop observing the
+                # flag (e.g. the process died between log lines).
+                job.status = JobStatus.STOPPED
+                buf.phase = "stopped"
+            elif exit_code == 0:
                 job.status = JobStatus.COMPLETED
                 buf.progress = 100.0
                 buf.phase = "done"
-                # Training outputs checkpoints, not video
-                ckpt_dirs = sorted(Path(job_output_dir).glob("checkpoint-*"))
+                # Training outputs checkpoints, not video. Sort by step number,
+                # not lexically ("checkpoint-1000" < "checkpoint-500" as strings).
+                ckpt_dirs = sorted(
+                    Path(job_output_dir).glob("checkpoint-*"),
+                    key=lambda p: int(m.group(1)) if (m := re.fullmatch(r"checkpoint-(\d+)", p.name)) else -1,
+                )
                 if ckpt_dirs:
                     job.output_path = str(ckpt_dirs[-1])
             else:
@@ -852,8 +1019,8 @@ class JobRunner:
 
     def _run_inference_job(self, job: Job):
         buf = job._log_buf
-        os.makedirs(self.log_dir, exist_ok=True)
-        job.log_file_path = os.path.join(self.log_dir, f"{job.id}.log")
+        os.makedirs(os.path.join(self.output_dir, job.id), exist_ok=True)
+        job.log_file_path = _job_log_path(self.output_dir, job.id)
 
         # Add file handler to persist logs
         file_handler = logging.FileHandler(job.log_file_path, mode='w', encoding='utf-8')
@@ -900,62 +1067,45 @@ class JobRunner:
             buf.phase = "loading model"
             logger.info("Loading model...")
 
-            # Run generator creation in a background thread so we
-            # can poll _stop_event while the (potentially slow)
-            # model download / load is in progress.
-            _gen_result: list[Any] = []
-            _gen_error: list[BaseException] = []
+            # The generator MUST be created on this thread: building it spawns
+            # the executor's worker processes, and they are torn down if the
+            # creating thread exits. Running it in a helper thread (to poll
+            # _stop_event during load) made every collective_rpc fail with
+            # ConnectionResetError.
+            if job._stop_event.is_set():
+                job.status = JobStatus.STOPPED
+                job.finished_at = time.time()
+                self._save_job(job)
+                logger.warning("Job %s stopped before model loading", job.id)
+                buf.phase = "stopped"
+                return
 
-            def _load_generator() -> None:
-                try:
-                    gen = self._get_or_create_generator(
-                        job.model_id,
-                        job.workload_type,
-                        job.num_gpus,
-                        dit_cpu_offload=job.dit_cpu_offload,
-                        text_encoder_cpu_offload=(job.text_encoder_cpu_offload),
-                        vae_cpu_offload=job.vae_cpu_offload,
-                        image_encoder_cpu_offload=(job.image_encoder_cpu_offload),
-                        use_fsdp_inference=job.use_fsdp_inference,
-                        enable_torch_compile=(job.enable_torch_compile),
-                        vsa_sparsity=job.vsa_sparsity,
-                        tp_size=job.tp_size,
-                        sp_size=job.sp_size,
-                        log_queue=log_queue,
-                    )
-                    _gen_result.append(gen)
-                except BaseException as exc:
-                    _gen_error.append(exc)
-
-            loader = threading.Thread(
-                target=_load_generator,
-                daemon=True,
+            generator = self._get_or_create_generator(
+                job.model_id,
+                job.workload_type,
+                job.num_gpus,
+                dit_cpu_offload=job.dit_cpu_offload,
+                dit_layerwise_offload=job.dit_layerwise_offload,
+                override_pipeline_cls_name=(MINIMAX_H3_REF2VA_PIPELINE if job.references else None),
+                text_encoder_cpu_offload=(job.text_encoder_cpu_offload),
+                vae_cpu_offload=job.vae_cpu_offload,
+                image_encoder_cpu_offload=(job.image_encoder_cpu_offload),
+                use_fsdp_inference=job.use_fsdp_inference,
+                enable_torch_compile=(job.enable_torch_compile),
+                vsa_sparsity=job.vsa_sparsity,
+                tp_size=job.tp_size,
+                sp_size=job.sp_size,
+                log_queue=log_queue,
             )
-            loader.start()
-
-            while loader.is_alive():
-                if job._stop_event.is_set():
-                    job.status = JobStatus.STOPPED
-                    job.finished_at = time.time()
-                    self._save_job(job)
-                    logger.warning(
-                        "Job %s stopped during model loading",
-                        job.id,
-                    )
-                    buf.phase = "stopped"
-                    return
-                loader.join(timeout=0.5)
-
-            if _gen_error:
-                raise _gen_error[0]
-
-            generator = _gen_result[0]
             buf.phase = "generating"
             logger.info("Starting generation for job %s (model=%s)", job.id, job.model_id)
 
+            # Without a name FastVideo derives the filename from the prompt.
+            safe_name = re.sub(r'[\\/:*?"<>|]+', "", job.name).strip().strip(".")
+            output_target = (os.path.join(job_output_dir, f"{safe_name[:80]}.mp4") if safe_name else job_output_dir)
             gen_kwargs: dict[str, Any] = {
                 "prompt": job.prompt,
-                "output_path": job_output_dir,
+                "output_path": output_target,
                 "save_video": True,
                 "num_inference_steps": job.num_inference_steps,
                 "num_frames": job.num_frames,
@@ -970,6 +1120,12 @@ class JobRunner:
             }
             if job.image_path:
                 gen_kwargs["image_path"] = job.image_path
+            if job.references:
+                gen_kwargs["references"] = _build_h3_references(job.references)
+            if job.last_image_path:
+                # _prepare_fl2va requires a PIL image, not a path.
+                from PIL import Image as _PILImage
+                gen_kwargs["last_image"] = _PILImage.open(job.last_image_path)
             generator.generate_video(**gen_kwargs)
 
             buf.phase = "saving"
@@ -1002,7 +1158,7 @@ class JobRunner:
 
         except Exception as exception:
             error_msg = str(exception)
-            logger.error("Critical error in job thread: %s", error_msg)
+            logger.exception("Critical error in job thread: %s", error_msg)
             job.status = JobStatus.FAILED
             job.error = f"Critical error ({type(exception).__name__}): {error_msg}"
             job.finished_at = time.time()

apps/fastvideo_studio/mock_server.py

@@ -0,0 +1,660 @@
+# SPDX-License-Identifier: Apache-2.0
+"""
+In-memory mock of the FastVideo Studio API (``server.py``).
+
+This mock implements the same ``/api`` routes and response shapes as the real
+FastAPI server but keeps everything in memory and never touches the FastVideo
+library, a GPU, or a database. It is meant to back the Playwright e2e suite so
+the Next.js frontend can be exercised end-to-end without a real backend.
+
+Job lifecycle is simulated by recording a start timestamp and *computing* the
+status on read: a started job reports ``running`` for a few seconds and then
+flips to ``completed`` with an ``output_path``, so polling the job list / logs
+shows progression. Generated media is a tiny 1s ``testsrc`` MP4 built lazily
+with ffmpeg and cached.
+
+Usage (from the ``apps/`` directory)::
+
+    PYTHONPATH=.. python -m fastvideo_studio.mock_server --port 8190
+"""
+
+from __future__ import annotations
+
+import contextlib
+import os
+import random
+import shutil
+import subprocess
+import tempfile
+import threading
+import time
+import uuid
+from typing import Annotated, Any
+
+from fastapi import FastAPI, File, HTTPException, UploadFile
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import FileResponse, PlainTextResponse
+
+from fastvideo_studio.database import default_settings_dict
+from fastvideo_studio.models import (CreateDatasetRequest, CreateJobRequest, SettingsUpdate, UpdateCaptionRequest,
+                                     model_label)
+
+# --- Config -----------------------------------------------------------------
+
+# How long a started job stays "running" before it flips to "completed".
+COMPLETE_AFTER_SECONDS = 3.0
+FFMPEG_BIN = shutil.which(os.getenv("FASTVIDEO_FFMPEG_BIN", "ffmpeg"))
+
+# A small catalogue of fake models keyed by workload type. Mirrors the real
+# server's {id, label} shape (label derived from the HF-style path).
+_MODELS_BY_WORKLOAD: dict[str, list[str]] = {
+    "t2v": [
+        "Wan-AI/Wan2.1-T2V-1.3B-Diffusers",
+        "FastVideo/FastHunyuan-diffusers",
+    ],
+    "i2v": [
+        "Wan-AI/Wan2.1-I2V-14B-480P-Diffusers",
+    ],
+    "t2i": [
+        "black-forest-labs/FLUX.1-schnell",
+    ],
+}
+
+
+def _models_for(workload_type: str | None) -> list[dict[str, str]]:
+    if workload_type:
+        paths = _MODELS_BY_WORKLOAD.get(workload_type, [])
+    else:
+        seen: dict[str, None] = {}
+        for paths_for_workload in _MODELS_BY_WORKLOAD.values():
+            for path in paths_for_workload:
+                seen.setdefault(path, None)
+        paths = list(seen)
+    return [{"id": path, "label": model_label(path)} for path in paths]
+
+
+# The real settings catalogue lives in database.py; the mock only pre-fills
+# the default model ids so the Create Job modal auto-selects one.
+_DEFAULT_SETTINGS: dict[str, Any] = {
+    **default_settings_dict(),
+    "defaultModelId": "Wan-AI/Wan2.1-T2V-1.3B-Diffusers",
+    "defaultModelIdT2v": "Wan-AI/Wan2.1-T2V-1.3B-Diffusers",
+    "defaultModelIdI2v": "Wan-AI/Wan2.1-I2V-14B-480P-Diffusers",
+    "defaultModelIdT2i": "black-forest-labs/FLUX.1-schnell",
+}
+
+# --- In-memory state --------------------------------------------------------
+
+_state_lock = threading.Lock()
+_settings: dict[str, Any] = dict(_DEFAULT_SETTINGS)
+_jobs: dict[str, dict[str, Any]] = {}
+_datasets: dict[str, dict[str, Any]] = {}
+# dataset_id -> {"file_names": [...], "captions": {file_name: caption}}
+_dataset_files: dict[str, dict[str, Any]] = {}
+
+# Lazily-built, cached media clips (shared by every job/dataset media response),
+# keyed by extension: "mp4" (1s testsrc video) or "png" (single testsrc frame).
+_mock_media_cache: dict[str, str] = {}
+_mock_media_lock = threading.Lock()
+
+
+def _build_mock_media(kind: str) -> str:
+    """Build (once) a tiny testsrc clip of the given ``kind`` with ffmpeg; cache the path."""
+    with _mock_media_lock:
+        cached = _mock_media_cache.get(kind)
+        if cached and os.path.isfile(cached):
+            return cached
+        if not FFMPEG_BIN:
+            raise HTTPException(status_code=500, detail="ffmpeg is required to build mock media")
+        fd, path = tempfile.mkstemp(prefix="fvstudio_mock_", suffix=f".{kind}")
+        os.close(fd)
+        if kind == "png":
+            command = [
+                FFMPEG_BIN,
+                "-hide_banner",
+                "-loglevel",
+                "error",
+                "-y",
+                "-f",
+                "lavfi",
+                "-i",
+                "testsrc=size=320x240:rate=1",
+                "-frames:v",
+                "1",
+                "-f",
+                "image2",
+                path,
+            ]
+        else:
+            command = [
+                FFMPEG_BIN,
+                "-hide_banner",
+                "-loglevel",
+                "error",
+                "-y",
+                "-f",
+                "lavfi",
+                "-i",
+                "testsrc=size=320x240:rate=24",
+                "-t",
+                "1",
+                "-c:v",
+                "libx264",
+                "-preset",
+                "ultrafast",
+                "-pix_fmt",
+                "yuv420p",
+                "-movflags",
+                "+faststart",
+                "-f",
+                "mp4",
+                path,
+            ]
+        try:
+            subprocess.run(command, check=True, capture_output=True)
+        except (subprocess.CalledProcessError, OSError) as exc:
+            with contextlib.suppress(OSError):
+                os.remove(path)
+            raise HTTPException(status_code=500, detail=f"ffmpeg failed to build mock {kind}: {exc}") from exc
+        if not os.path.isfile(path) or os.path.getsize(path) == 0:
+            raise HTTPException(status_code=500, detail=f"ffmpeg produced no mock {kind} bytes")
+        _mock_media_cache[kind] = path
+        return path
+
+
+# --- Job helpers ------------------------------------------------------------
+
+
+def _new_job_dict(req: CreateJobRequest) -> dict[str, Any]:
+    """Build a job dict (mirrors job_runner.Job.to_dict()) in the pending state."""
+    job_id = str(uuid.uuid4())
+    return {
+        "id": job_id,
+        "model_id": req.model_id,
+        "prompt": req.prompt,
+        "workload_type": req.workload_type or "t2v",
+        "job_type": req.job_type or "inference",
+        "image_path": req.image_path or "",
+        "status": "pending",
+        "created_at": time.time(),
+        "started_at": None,
+        "finished_at": None,
+        "error": None,
+        "output_path": None,
+        "log_file_path": None,
+        "num_inference_steps": req.num_inference_steps,
+        "num_frames": req.num_frames,
+        "height": req.height,
+        "width": req.width,
+        "guidance_scale": req.guidance_scale,
+        "guidance_rescale": req.guidance_rescale,
+        "fps": req.fps,
+        "seed": req.seed,
+        "negative_prompt": req.negative_prompt or "",
+        "num_gpus": req.num_gpus,
+        "data_path": req.data_path or "",
+        "progress": 0.0,
+        "progress_msg": "",
+        "phase": "pending",
+    }
+
+
+def _advance_job(job: dict[str, Any]) -> None:
+    """Flip a running job to completed once enough wall-clock time has passed.
+
+    Status is *computed on read* from the recorded start timestamp, so polling
+    the job list / logs naturally shows pending -> running -> completed.
+    """
+    if job["status"] != "running" or not job.get("started_at"):
+        return
+    elapsed = time.time() - job["started_at"]
+    if elapsed >= COMPLETE_AFTER_SECONDS:
+        job["status"] = "completed"
+        job["finished_at"] = job["started_at"] + COMPLETE_AFTER_SECONDS
+        job["progress"] = 100.0
+        job["progress_msg"] = "50/50 steps"
+        job["phase"] = "done"
+        ext = "png" if job.get("workload_type") == "t2i" else "mp4"
+        job["output_path"] = f"/mock/outputs/{job['id']}/output.{ext}"
+
+
+def _public_job(job: dict[str, Any]) -> dict[str, Any]:
+    """Advance + return a copy safe to serialize."""
+    _advance_job(job)
+    return dict(job)
+
+
+_LOG_TAIL = [
+    "Loading model...",
+    "Model loaded.",
+    "Starting generation...",
+    "Denoising step 10/50",
+    "Denoising step 20/50",
+    "Denoising step 30/50",
+    "Denoising step 40/50",
+    "Denoising step 50/50",
+    "Generation complete. Saving output...",
+    "Saved output file.",
+    "Job completed successfully.",
+]
+
+
+def _log_sequence(job: dict[str, Any]) -> list[str]:
+    return [
+        f"Job {job['id']} started",
+        f"Model: {job['model_id']}",
+        f"Prompt: {job['prompt']}",
+        *_LOG_TAIL,
+    ]
+
+
+def _compute_logs(job: dict[str, Any]) -> dict[str, Any]:
+    """Return JobLogs-shaped data, growing the visible lines as time passes."""
+    seq = _log_sequence(job)
+    status = job["status"]
+    if status == "pending":
+        return {"lines": [], "progress": 0.0, "progress_msg": "", "phase": "pending"}
+    if status == "completed":
+        return {"lines": seq, "progress": 100.0, "progress_msg": "50/50 steps", "phase": "done"}
+    if status in ("failed", "stopped"):
+        # Keep the total line count monotonic across running -> terminal (the
+        # `after` cursor relies on it): show every non-final line plus a terminal
+        # notice, which is always >= any prefix a running job revealed.
+        lines = seq[:-1] + [f"Job {status}."]
+        return {"lines": lines, "progress": job.get("progress", 0.0), "progress_msg": "", "phase": status}
+    # running: reveal a prefix proportional to elapsed time, but never the final
+    # "completed" line — that appears only once the job is actually completed.
+    elapsed = time.time() - (job.get("started_at") or time.time())
+    frac = max(0.0, min(elapsed / COMPLETE_AFTER_SECONDS, 0.99))
+    reveal = min(max(4, round(frac * len(seq))), len(seq) - 1)
+    lines = seq[:reveal]
+    if frac < 0.2:
+        phase = "loading model"
+    elif frac < 0.85:
+        phase = "denoising"
+    else:
+        phase = "saving"
+    return {
+        "lines": lines,
+        "progress": round(frac * 100.0, 1),
+        "progress_msg": f"{int(frac * 50)}/50 steps",
+        "phase": phase,
+    }
+
+
+# --- Dataset helpers --------------------------------------------------------
+
+
+def _dataset_stats(dataset_id: str) -> tuple[int, int]:
+    files = _dataset_files.get(dataset_id, {}).get("file_names", [])
+    count = len(files)
+    # Fake but stable per-file size so the UI shows a non-zero footprint.
+    return count, count * 1_048_576
+
+
+def _public_dataset(dataset: dict[str, Any]) -> dict[str, Any]:
+    count, size = _dataset_stats(dataset["id"])
+    return {**dataset, "file_count": count, "size_bytes": size}
+
+
+def _seed() -> None:
+    """Seed a couple of datasets and one completed inference job."""
+    now = time.time()
+    seeds = [
+        ("Sunset Clips", ["sunset_01.mp4", "sunset_02.mp4"], {
+            "sunset_01.mp4": "A sunset over the ocean",
+            "sunset_02.mp4": "A sunset over the mountains"
+        }),
+        ("City Timelapse", ["city_01.mp4"], {
+            "city_01.mp4": "A busy city intersection at night"
+        }),
+    ]
+    for idx, (name, file_names, captions) in enumerate(seeds):
+        dataset_id = str(uuid.uuid4())
+        _datasets[dataset_id] = {"id": dataset_id, "name": name, "created_at": now - 600 + idx}
+        _dataset_files[dataset_id] = {"file_names": list(file_names), "captions": dict(captions)}
+
+    job_id = str(uuid.uuid4())
+    _jobs[job_id] = {
+        "id": job_id,
+        "model_id": "Wan-AI/Wan2.1-T2V-1.3B-Diffusers",
+        "prompt": "A curious raccoon peers through a field of yellow sunflowers",
+        "workload_type": "t2v",
+        "job_type": "inference",
+        "image_path": "",
+        "status": "completed",
+        "created_at": now - 300,
+        "started_at": now - 280,
+        "finished_at": now - 250,
+        "error": None,
+        "output_path": f"/mock/outputs/{job_id}/output.mp4",
+        "log_file_path": f"/mock/logs/{job_id}.log",
+        "num_inference_steps": 50,
+        "num_frames": 81,
+        "height": 480,
+        "width": 832,
+        "guidance_scale": 5.0,
+        "guidance_rescale": 0.0,
+        "fps": 24,
+        "seed": 1024,
+        "negative_prompt": "",
+        "num_gpus": 1,
+        "data_path": "",
+        "progress": 100.0,
+        "progress_msg": "50/50 steps",
+        "phase": "done",
+    }
+
+
+# --- App --------------------------------------------------------------------
+
+app = FastAPI(title="FastVideo Studio Mock API", version="0.1.0")
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+_seed()
+
+
+@app.get("/api/__mock__")
+def mock_sentinel() -> dict[str, bool]:
+    """Mock-only marker so the e2e suite can prove it isn't hitting a real
+    backend before running its mutating specs (the real server has no such
+    route)."""
+    return {"mock": True}
+
+
+# --- Settings ---------------------------------------------------------------
+
+
+@app.get("/api/settings")
+def get_settings() -> dict[str, Any]:
+    with _state_lock:
+        return dict(_settings)
+
+
+@app.put("/api/settings")
+def update_settings(settings: SettingsUpdate) -> dict[str, Any]:
+    updates = settings.model_dump(exclude_unset=True)
+    with _state_lock:
+        _settings.update({k: v for k, v in updates.items() if v is not None})
+        return dict(_settings)
+
+
+# --- Models -----------------------------------------------------------------
+
+
+@app.get("/api/models")
+def list_models(workload_type: str | None = None) -> list[dict[str, Any]]:
+    return _models_for(workload_type)
+
+
+# --- GPUs -------------------------------------------------------------------
+
+
+@app.get("/api/gpus")
+def list_gpus() -> dict[str, Any]:
+    """Two fake GPUs with slight per-request jitter so the page looks live."""
+    gpus = []
+    for index, (base_util, used_mib) in enumerate([(62, 61_440), (7, 4_096)]):
+        gpus.append({
+            "index": index,
+            "name": "NVIDIA Mock GPU 80GB",
+            "utilization": max(0, min(100, base_util + random.randint(-5, 5))),
+            "memory_used_mib": used_mib + random.randint(-256, 256),
+            "memory_total_mib": 81_920,
+            "temperature_c": 55 + random.randint(-3, 3),
+            "power_watts": 310.0 + random.randint(-20, 20),
+            "power_limit_watts": 700.0,
+        })
+    return {"available": True, "gpus": gpus, "error": None}
+
+
+# --- Uploads ----------------------------------------------------------------
+
+
+@app.post("/api/upload-image")
+async def upload_image(file: Annotated[UploadFile, File()]) -> dict[str, str]:
+    name = file.filename or "image.png"
+    return {"path": f"/mock/uploads/{uuid.uuid4().hex}_{os.path.basename(name)}"}
+
+
+@app.post("/api/upload-raw-dataset")
+async def upload_raw_dataset(files: Annotated[list[UploadFile], File()]) -> dict[str, Any]:
+    video_exts = {".mp4", ".webm", ".avi", ".mov", ".mkv"}
+    file_names: list[str] = []
+    for uf in files:
+        name = os.path.basename(uf.filename or f"{uuid.uuid4().hex}.mp4")
+        if os.path.splitext(name)[1].lower() in video_exts:
+            file_names.append(name)
+    if not file_names:
+        raise HTTPException(status_code=400, detail="No video files found.")
+    upload_id = uuid.uuid4().hex
+    path = f"/mock/uploads/{upload_id}"
+    return {"path": path, "upload_id": upload_id, "file_names": file_names}
+
+
+# --- Jobs -------------------------------------------------------------------
+
+
+@app.get("/api/jobs")
+def list_jobs(job_type: str | None = None) -> list[dict[str, Any]]:
+    # _public_job mutates the shared job dict via _advance_job, so it must run
+    # under the lock (matching every other job route) to avoid racing start/stop.
+    with _state_lock:
+        jobs = list(_jobs.values())
+        if job_type:
+            jobs = [j for j in jobs if j.get("job_type") == job_type]
+        jobs.sort(key=lambda j: j["created_at"], reverse=True)
+        return [_public_job(j) for j in jobs]
+
+
+@app.get("/api/jobs/{job_id}")
+def get_job(job_id: str) -> dict[str, Any]:
+    with _state_lock:
+        job = _jobs.get(job_id)
+        if job is None:
+            raise HTTPException(status_code=404, detail="Job not found")
+        return _public_job(job)
+
+
+@app.post("/api/jobs", status_code=201)
+def create_job(req: CreateJobRequest) -> dict[str, Any]:
+    job = _new_job_dict(req)
+    with _state_lock:
+        _jobs[job["id"]] = job
+        if _settings.get("autoStartJob"):
+            _start(job)
+        return _public_job(job)
+
+
+def _start(job: dict[str, Any]) -> None:
+    job["status"] = "running"
+    job["started_at"] = time.time()
+    job["finished_at"] = None
+    job["error"] = None
+    job["output_path"] = None
+    # The real server assigns the log path once the job starts running; mirror
+    # that so the Job Details "Download Log" button (gated on log_file_path) works.
+    job["log_file_path"] = f"/mock/logs/{job['id']}.log"
+    job["progress"] = 0.0
+    job["progress_msg"] = ""
+    job["phase"] = "starting"
+
+
+@app.post("/api/jobs/{job_id}/start")
+def start_job(job_id: str) -> dict[str, Any]:
+    with _state_lock:
+        job = _jobs.get(job_id)
+        if job is None:
+            raise HTTPException(status_code=404, detail=f"Job {job_id} not found")
+        _advance_job(job)
+        if job["status"] == "running":
+            raise HTTPException(status_code=409, detail="Job is already running")
+        if job["status"] == "completed":
+            raise HTTPException(status_code=409, detail="Job already completed. Delete and re-create to run again.")
+        _start(job)
+        return _public_job(job)
+
+
+@app.post("/api/jobs/{job_id}/stop")
+def stop_job(job_id: str) -> dict[str, Any]:
+    with _state_lock:
+        job = _jobs.get(job_id)
+        if job is None:
+            raise HTTPException(status_code=404, detail=f"Job {job_id} not found")
+        _advance_job(job)
+        if job["status"] != "running":
+            raise HTTPException(status_code=409, detail=f"Job is not running (status={job['status']})")
+        job["status"] = "stopped"
+        job["finished_at"] = time.time()
+        job["phase"] = "stopped"
+        return dict(job)
+
+
+@app.delete("/api/jobs/{job_id}")
+def delete_job(job_id: str) -> dict[str, str]:
+    with _state_lock:
+        if _jobs.pop(job_id, None) is None:
+            raise HTTPException(status_code=404, detail="Job not found")
+    return {"detail": f"Job {job_id} deleted"}
+
+
+@app.get("/api/jobs/{job_id}/logs")
+def get_job_logs(job_id: str, after: int = 0) -> dict[str, Any]:
+    with _state_lock:
+        job = _jobs.get(job_id)
+        if job is None:
+            raise HTTPException(status_code=404, detail=f"Job {job_id} not found")
+        _advance_job(job)
+        result = _compute_logs(job)
+    all_lines = result["lines"]
+    return {
+        "lines": all_lines[after:],
+        "total": len(all_lines),
+        "progress": result["progress"],
+        "progress_msg": result["progress_msg"],
+        "phase": result["phase"],
+    }
+
+
+@app.get("/api/jobs/{job_id}/video")
+def get_video(job_id: str) -> FileResponse:
+    with _state_lock:
+        job = _jobs.get(job_id)
+        if job is None:
+            raise HTTPException(status_code=404, detail="Job not found")
+        _advance_job(job)
+        output_path = job.get("output_path")
+        if job["status"] != "completed" or not output_path:
+            raise HTTPException(status_code=404, detail="No output available for this job")
+    # Mirror the real server: image workloads (t2i) output a .png served as an
+    # image; everything else is a video.
+    if output_path.endswith(".png"):
+        return FileResponse(_build_mock_media("png"), media_type="image/png", filename=f"job_{job_id}.png")
+    return FileResponse(_build_mock_media("mp4"), media_type="video/mp4", filename=f"job_{job_id}.mp4")
+
+
+@app.get("/api/jobs/{job_id}/download_log")
+def download_log(job_id: str) -> PlainTextResponse:
+    with _state_lock:
+        job = _jobs.get(job_id)
+        if job is None:
+            raise HTTPException(status_code=404, detail="Job not found")
+        _advance_job(job)
+        lines = _compute_logs(job)["lines"]
+    return PlainTextResponse("\n".join(lines) + "\n", media_type="text/plain")
+
+
+# --- Datasets ---------------------------------------------------------------
+
+
+@app.get("/api/datasets")
+def list_datasets() -> list[dict[str, Any]]:
+    with _state_lock:
+        datasets = sorted(_datasets.values(), key=lambda d: d["created_at"], reverse=True)
+        return [_public_dataset(d) for d in datasets]
+
+
+@app.get("/api/datasets/{dataset_id}")
+def get_dataset(dataset_id: str) -> dict[str, Any]:
+    with _state_lock:
+        dataset = _datasets.get(dataset_id)
+        if dataset is None:
+            raise HTTPException(status_code=404, detail="Dataset not found")
+        return _public_dataset(dataset)
+
+
+@app.post("/api/datasets", status_code=201)
+def create_dataset(req: CreateDatasetRequest) -> dict[str, Any]:
+    if not req.upload_path:
+        raise HTTPException(status_code=400, detail="upload_path is required. Upload media files first.")
+    if not req.file_names:
+        raise HTTPException(status_code=400, detail="No media files found.")
+    dataset_id = str(uuid.uuid4())
+    dataset = {"id": dataset_id, "name": req.name, "created_at": time.time()}
+    captions = {fn: (req.captions.get(fn, "") if req.captions else "") for fn in req.file_names}
+    with _state_lock:
+        _datasets[dataset_id] = dataset
+        _dataset_files[dataset_id] = {"file_names": list(req.file_names), "captions": captions}
+        return _public_dataset(dataset)
+
+
+@app.get("/api/datasets/{dataset_id}/files")
+def get_dataset_files(dataset_id: str) -> dict[str, Any]:
+    with _state_lock:
+        if dataset_id not in _datasets:
+            raise HTTPException(status_code=404, detail="Dataset not found")
+        files = _dataset_files.get(dataset_id, {"file_names": [], "captions": {}})
+        return {"file_names": list(files["file_names"]), "captions": dict(files["captions"])}
+
+
+@app.put("/api/datasets/{dataset_id}/captions")
+def update_dataset_caption(dataset_id: str, req: UpdateCaptionRequest) -> dict[str, str]:
+    with _state_lock:
+        if dataset_id not in _datasets:
+            raise HTTPException(status_code=404, detail="Dataset not found")
+        files = _dataset_files.setdefault(dataset_id, {"file_names": [], "captions": {}})
+        files["captions"][req.file_name] = req.caption
+    return {"detail": "Caption updated"}
+
+
+@app.get("/api/datasets/{dataset_id}/media/{file_name:path}")
+def serve_dataset_media(dataset_id: str, file_name: str) -> FileResponse:
+    with _state_lock:
+        if dataset_id not in _datasets:
+            raise HTTPException(status_code=404, detail="Dataset not found")
+        if file_name not in _dataset_files.get(dataset_id, {}).get("file_names", []):
+            raise HTTPException(status_code=404, detail="File not found")
+    return FileResponse(_build_mock_media("mp4"), media_type="video/mp4")
+
+
+@app.delete("/api/datasets/{dataset_id}")
+def delete_dataset(dataset_id: str) -> dict[str, str]:
+    with _state_lock:
+        if _datasets.pop(dataset_id, None) is None:
+            raise HTTPException(status_code=404, detail="Dataset not found")
+        _dataset_files.pop(dataset_id, None)
+    return {"detail": f"Dataset {dataset_id} deleted"}
+
+
+def main() -> None:
+    import argparse
+
+    import uvicorn
+
+    parser = argparse.ArgumentParser(description="FastVideo Studio mock API server")
+    parser.add_argument("--host", default="127.0.0.1", help="Bind address (default: 127.0.0.1)")
+    # Default off the real server's 8189 so the mock never shadows a real API.
+    parser.add_argument("--port", type=int, default=8190, help="Port number (default: 8190)")
+    args = parser.parse_args()
+
+    uvicorn.run(app, host=args.host, port=args.port, log_level="info")
+
+
+if __name__ == "__main__":
+    main()

apps/fastvideo_studio/models/__init__.py

@@ -1,14 +1,22 @@
 # SPDX-License-Identifier: Apache-2.0
-"""Pydantic request/response models for the API."""
+"""Pydantic request/response models for the API, plus tiny shared helpers
+usable by both the real server and the dependency-light mock server."""
 
 from fastvideo_studio.models.create_job_request import CreateJobRequest
 from fastvideo_studio.models.settings_update import SettingsUpdate
 from fastvideo_studio.models.create_dataset_request import CreateDatasetRequest
 from fastvideo_studio.models.update_caption_request import UpdateCaptionRequest
 
+
+def model_label(model_path: str) -> str:
+    """Derive a readable label from an HF-style model path."""
+    return model_path.split("/")[-1].replace("-", " ").replace("_", " ")
+
+
 __all__ = [
     "CreateJobRequest",
     "SettingsUpdate",
     "CreateDatasetRequest",
     "UpdateCaptionRequest",
+    "model_label",
 ]

apps/fastvideo_studio/models/create_job_request.py

@@ -1,15 +1,20 @@
 # SPDX-License-Identifier: Apache-2.0
 """Request model for creating a job."""
 
+from typing import Any
+
 from pydantic import BaseModel
 
 
 class CreateJobRequest(BaseModel):
     model_id: str
+    name: str = ""
     prompt: str
     workload_type: str = "t2v"
     job_type: str = "inference"
     image_path: str = ""
+    last_image_path: str = ""
+    references: list[dict[str, Any]] | None = None
     data_path: str = ""
     max_train_steps: int = 1000
     train_batch_size: int = 1
@@ -17,7 +22,6 @@ class CreateJobRequest(BaseModel):
     num_latent_t: int = 20
     validation_dataset_file: str = ""
     lora_rank: int = 32
-    ltx2_first_frame_conditioning_p: float | None = None
     negative_prompt: str = ""
     num_inference_steps: int = 50
     num_frames: int = 81
@@ -29,6 +33,7 @@ class CreateJobRequest(BaseModel):
     seed: int = 1024
     num_gpus: int = 1
     dit_cpu_offload: bool = False
+    dit_layerwise_offload: bool = False
     text_encoder_cpu_offload: bool = False
     vae_cpu_offload: bool = False
     image_encoder_cpu_offload: bool = False
@@ -41,8 +46,6 @@ class CreateJobRequest(BaseModel):
     dmd_use_vsa: bool = False
     dmd_vsa_sparsity: float = 0.8
     dmd_denoising_steps: str = "1000,757,522"
-    min_timestep_ratio: float = 0.02
-    max_timestep_ratio: float = 0.98
     real_score_guidance_scale: float = 3.5
     generator_update_interval: int = 5
     real_score_model_path: str = ""

apps/fastvideo_studio/next.config.ts

@@ -0,0 +1,13 @@
+import path from 'node:path';
+import { fileURLToPath } from 'node:url';
+import type { NextConfig } from 'next';
+
+const configDir = path.dirname(fileURLToPath(import.meta.url));
+
+const nextConfig: NextConfig = {
+  // Point tracing at the monorepo root (apps/fastvideo_studio -> apps -> repo
+  // root) so Next stops warning about multiple lockfiles in the workspace.
+  outputFileTracingRoot: path.join(configDir, '..', '..'),
+};
+
+export default nextConfig;