[bugfix]: add the missing DiT actor shutdown and LoRA RPCs for disaggregated H3

RayMiniMaxH3DisaggregatedRuntime.close() called shutdown on both actors,
but _MiniMaxH3DiTActor defined no shutdown method, so every explicit
close raised AttributeError after the actors were already killed. The
executor also forwarded set_lora_adapter / unmerge_lora_weights /
merge_lora_weights to the DiT actor, which implemented none of them.

- Move health() and shutdown() into a shared _MiniMaxH3ResidentActor
  base so both roles expose the same lifecycle RPCs.
- Add the three LoRA RPCs to the DiT actor, delegating to its resident
  pipeline and returning the same receipts as GPU workers; the executor
  now checks the receipt for all three calls, not only set.
- Log per-request [RAY_PAYLOAD] sizes only when
  FASTVIDEO_H3_PROFILE_TRANSFERS is set, like the other transfer logs.
- Tests use a Ray-faithful actor handle that only exposes methods the
  real actor class defines; the new tests fail on the previous head.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

.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/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/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 +1,52 @@
+#!/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
+
+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"
 }

.buildkite/scripts/unit_test.sh

@@ -0,0 +1,25 @@
+#!/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/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

@@ -26,29 +26,48 @@ 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 +75,13 @@ 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 +89,6 @@ 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-macos-mlx.yml

@@ -8,6 +8,8 @@ on:
       - "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"
@@ -78,6 +80,13 @@ jobs:
             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 \
@@ -90,7 +99,8 @@ jobs:
             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 \
-            -q
+            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
@@ -134,6 +144,13 @@ jobs:
             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 \
@@ -146,4 +163,5 @@ jobs:
             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 \
-            -q
+            fastvideo/tests/platforms/test_cpu_sdpa.py \
+            -v -s -o faulthandler_timeout=120

.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 golden-gate 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
@@ -137,7 +95,17 @@ jobs:
 
           declare -A MAP=(
             [encoder]=encoder [vae]=vae [transformer]=transformer
-            [kernel]=kernel_tests [unit]=unit_test [dreamverse]=dreamverse_app
+            [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

.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

@@ -13,6 +13,11 @@ on:
         required: false
         default: false
         type: boolean
+      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.
@@ -198,6 +203,28 @@ 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_100 rather than sm_121.
+  # 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=10.0
+        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/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
@@ -168,7 +169,8 @@ 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
@@ -178,7 +180,7 @@ jobs:
             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;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

@@ -55,6 +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/
@@ -133,6 +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/.*|

README.md

@@ -3,12 +3,14 @@
 </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
+- `2026/09/01`: 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 [Apple Silicon guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/mps/) 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/).
@@ -63,9 +65,10 @@ 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/).
 
-> **On an Apple Silicon Mac?** FastVideo runs FastWan text-to-video natively
-> through an MLX runtime — a 5-second 480p clip generated locally, no cloud,
-> no discrete GPU. Install with `uv pip install -e '.[mlx]'` and follow the
+> **On an Apple Silicon Mac?** FastVideo runs FastMetal-QAD through an MLX
+> runtime. Install with `uv pip install -e '.[mlx]'`, download
+> [`FastVideo/FastMetal-1.3B-QAD`](https://huggingface.co/FastVideo/FastMetal-1.3B-QAD),
+> and follow the
 > [Apple Silicon guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/mps/).
 
 Please see our [docs](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/) for more detailed installation instructions.

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/test_config.py

@@ -2,13 +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",
@@ -20,7 +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")
 
@@ -150,3 +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_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

@@ -63,6 +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")
@@ -71,6 +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
@@ -92,9 +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

@@ -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",
     ),
 }

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/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/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,
@@ -254,14 +264,17 @@ class Database:
                 dmd_use_vsa, dmd_vsa_sparsity, dmd_denoising_steps,
                 real_score_guidance_scale,
                 generator_update_interval, real_score_model_path, fake_score_model_path
-            ) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
+            ) 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"],
@@ -540,9 +553,12 @@ def _row_to_job(row: sqlite3.Row) -> dict[str, Any]:
     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"],

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
@@ -123,10 +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
@@ -145,6 +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
@@ -180,10 +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,
@@ -202,6 +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,
@@ -232,6 +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."""
 
@@ -276,7 +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:
@@ -314,10 +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),
@@ -350,6 +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),
@@ -394,9 +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,
@@ -422,6 +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,
@@ -435,10 +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,
@@ -464,6 +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,
@@ -521,6 +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.
         
@@ -623,6 +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,
@@ -638,6 +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,
@@ -650,8 +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
@@ -677,6 +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,
@@ -706,10 +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
@@ -827,8 +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')
@@ -875,62 +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,
@@ -945,6 +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"
@@ -977,7 +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/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
@@ -28,6 +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

apps/fastvideo_studio/server.py

@@ -18,6 +18,7 @@ import argparse
 import contextlib
 import logging
 import os
+import re
 import shutil
 import signal
 import time
@@ -116,7 +117,30 @@ def list_models(workload_type: str | None = None) -> list[dict[str, Any]]:
     return _available_models
 
 
+def _safe_upload_name(filename: str | None, ext: str) -> str:
+    """A filesystem-safe version of the client's filename, keeping it readable.
+
+    Uploads live under a per-file uuid directory, so the basename does not have
+    to be unique -- only safe. Keeping the original name means the path stays
+    self-describing wherever it travels: the database, job logs, and payloads
+    copied back out to the API.
+    """
+    stem = os.path.basename(filename or "").rsplit(".", 1)[0]
+    stem = re.sub(r"[^A-Za-z0-9._-]+", "_", stem).strip("._-")
+    return f"{stem[:80] or 'upload'}{ext}"
+
+
+def _upload_destination(ext: str, filename: str | None) -> str:
+    """<upload_dir>/<uuid4>/<safe original name><ext>"""
+    directory = os.path.join(upload_dir, uuid.uuid4().hex)
+    os.makedirs(directory, exist_ok=True)
+    return os.path.join(directory, _safe_upload_name(filename, ext))
+
+
 ALLOWED_IMAGE_EXTENSIONS = {".png", ".jpg", ".jpeg", ".webp", ".bmp"}
+ALLOWED_VIDEO_EXTENSIONS = {".mp4", ".mov", ".mkv", ".webm", ".avi"}
+ALLOWED_AUDIO_EXTENSIONS = {".wav", ".mp3", ".flac", ".m4a", ".ogg"}
+ALLOWED_MEDIA_EXTENSIONS = (ALLOWED_IMAGE_EXTENSIONS | ALLOWED_VIDEO_EXTENSIONS | ALLOWED_AUDIO_EXTENSIONS)
 
 
 @app.post("/api/upload-image")
@@ -136,8 +160,7 @@ async def upload_image(file: Annotated[UploadFile, File()], ) -> dict[str, str]:
                     f"{', '.join(ALLOWED_IMAGE_EXTENSIONS)}"),
         )
     os.makedirs(upload_dir, exist_ok=True)
-    unique_name = f"{uuid.uuid4().hex}{ext}"
-    dest_path = os.path.join(upload_dir, unique_name)
+    dest_path = _upload_destination(ext, file.filename)
     try:
         contents = await file.read()
         with open(dest_path, "wb") as f:
@@ -150,6 +173,47 @@ async def upload_image(file: Annotated[UploadFile, File()], ) -> dict[str, str]:
     return {"path": os.path.abspath(dest_path)}
 
 
+@app.post("/api/upload-media")
+async def upload_media(file: Annotated[UploadFile, File()], ) -> dict[str, str]:
+    """Upload an image, video or audio file for Ref2VA references.
+
+    Returns the absolute path plus the media_type MiniMax-H3 expects, so the
+    caller does not have to re-derive it from the extension.
+    """
+    global upload_dir  # noqa: PLW0603
+    if not upload_dir:
+        raise HTTPException(
+            status_code=503,
+            detail="Upload directory not configured",
+        )
+    ext = Path(file.filename or "").suffix.lower()
+    if ext not in ALLOWED_MEDIA_EXTENSIONS:
+        raise HTTPException(
+            status_code=400,
+            detail=(f"Invalid file type. Allowed: "
+                    f"{', '.join(sorted(ALLOWED_MEDIA_EXTENSIONS))}"),
+        )
+    if ext in ALLOWED_VIDEO_EXTENSIONS:
+        media_type = "video"
+    elif ext in ALLOWED_AUDIO_EXTENSIONS:
+        media_type = "audio"
+    else:
+        media_type = "image"
+
+    os.makedirs(upload_dir, exist_ok=True)
+    dest_path = _upload_destination(ext, file.filename)
+    try:
+        contents = await file.read()
+        with open(dest_path, "wb") as f:
+            f.write(contents)
+    except OSError as e:
+        raise HTTPException(
+            status_code=500,
+            detail=f"Failed to save upload: {e}",
+        ) from e
+    return {"path": os.path.abspath(dest_path), "media_type": media_type}
+
+
 ALLOWED_VIDEO_EXTENSIONS = {".mp4", ".webm", ".avi", ".mov", ".mkv"}
 
 
@@ -282,10 +346,13 @@ def create_job(req: CreateJobRequest) -> dict[str, Any]:
     job = job_runner.create_job(
         job_id=str(uuid.uuid4()),
         model_id=req.model_id,
+        name=req.name or "",
         prompt=req.prompt,
         workload_type=req.workload_type or "t2v",
         job_type=job_type,
         image_path=req.image_path or "",
+        last_image_path=req.last_image_path or "",
+        references=req.references or [],
         data_path=data_path,
         max_train_steps=req.max_train_steps,
         train_batch_size=req.train_batch_size,
@@ -304,6 +371,7 @@ def create_job(req: CreateJobRequest) -> dict[str, Any]:
         seed=req.seed,
         num_gpus=req.num_gpus,
         dit_cpu_offload=req.dit_cpu_offload,
+        dit_layerwise_offload=req.dit_layerwise_offload,
         text_encoder_cpu_offload=req.text_encoder_cpu_offload,
         vae_cpu_offload=req.vae_cpu_offload,
         image_encoder_cpu_offload=req.image_encoder_cpu_offload,
@@ -336,6 +404,28 @@ def create_job(req: CreateJobRequest) -> dict[str, Any]:
     return job.to_dict()
 
 
+@app.post("/api/jobs/{job_id}/duplicate", status_code=201)
+def duplicate_job(job_id: str) -> dict[str, Any]:
+    """Create a new pending job with the same configuration as an existing one."""
+    try:
+        job = job_runner.duplicate_job(job_id, str(uuid.uuid4()))
+    except ValueError as e:
+        raise HTTPException(status_code=404, detail=str(e)) from e
+    return job.to_dict()
+
+
+@app.patch("/api/jobs/{job_id}")
+def update_job(job_id: str, updates: dict[str, Any]) -> dict[str, Any]:
+    """Edit a pending job's configuration. Started jobs cannot be edited."""
+    try:
+        job = job_runner.update_job_config(job_id, updates)
+    except ValueError as e:
+        detail = str(e)
+        status = 404 if "not found" in detail else 400
+        raise HTTPException(status_code=status, detail=detail) from e
+    return job.to_dict()
+
+
 @app.post("/api/jobs/{job_id}/start")
 def start_job(job_id: str) -> dict[str, Any]:
     """Start (or restart) a pending / stopped / failed job."""

apps/fastvideo_studio/src/components/jobs/CreateJobModal.tsx

@@ -22,15 +22,35 @@ import { useStore } from '@/hooks/useStore';
 import { defaultOptionsStore } from '@/stores/defaultOptions';
 import {
   createJob,
+  updateJob,
   getDatasets,
   getModels,
   uploadImage,
+  uploadMedia,
   type CreateJobRequest,
   type Model,
 } from '@/lib/api';
 import { getDefaultModelForWorkload } from '@/lib/defaultOptions';
 import { WORKLOAD_OPTIONS } from '@/lib/jobConfig';
 import type { JobType } from '@/lib/types';
+import {
+  H3_MAX_REFERENCES,
+  labelReferences,
+  referencePromptSeed,
+  validateReferences,
+  type H3Reference,
+} from '@/lib/h3References';
+import {
+  EMPTY_H3_PROMPT_FIELDS,
+  H3_PROMPT_SECTIONS,
+  H3_SECTION_HINTS,
+  H3_SECTION_LABELS,
+  isEmptyPromptFields,
+  parseH3Prompt,
+  serializeH3Prompt,
+  type H3PromptFields,
+} from '@/lib/h3Prompt';
+import { jobToFormFields, type JobLike } from '@/lib/jobToFields';
 
 export interface CreateJobModalProps {
   isOpen: boolean;
@@ -38,6 +58,10 @@ export interface CreateJobModalProps {
   onSuccess: () => void;
   jobType: JobType;
   workloadType: string;
+  /** When set, the modal edits this pending job instead of creating a new one. */
+  editingJob?: JobLike | null;
+  /** Show the configuration without allowing changes (started/finished jobs). */
+  readOnly?: boolean;
 }
 
 export default function CreateJobModal({
@@ -46,6 +70,8 @@ export default function CreateJobModal({
   onSuccess,
   jobType,
   workloadType,
+  editingJob,
+  readOnly = false,
 }: CreateJobModalProps) {
   const { options } = useStore(defaultOptionsStore);
 
@@ -55,8 +81,22 @@ export default function CreateJobModal({
 
   const [models, setModels] = React.useState<Model[]>([]);
   const [modelId, setModelId] = React.useState('');
+  const [name, setName] = React.useState('');
   const [prompt, setPrompt] = React.useState('');
   const [imagePath, setImagePath] = React.useState('');
+  const [lastImagePath, setLastImagePath] = React.useState('');
+  const [references, setReferences] = React.useState<H3Reference[]>([]);
+  const [isUploadingReference, setIsUploadingReference] = React.useState(false);
+  const [referenceError, setReferenceError] = React.useState<string | null>(null);
+  const [promptFields, setPromptFields] = React.useState<H3PromptFields>(
+    EMPTY_H3_PROMPT_FIELDS,
+  );
+  const [useGuidedPrompt, setUseGuidedPrompt] = React.useState(true);
+  const [lastImageFileName, setLastImageFileName] = React.useState('');
+  const [isUploadingLastImage, setIsUploadingLastImage] = React.useState(false);
+  const [lastImageUploadError, setLastImageUploadError] = React.useState<
+    string | null
+  >(null);
   const [imageFileName, setImageFileName] = React.useState('');
   const [isUploadingImage, setIsUploadingImage] = React.useState(false);
   const [negativePrompt, setNegativePrompt] = React.useState('');
@@ -70,12 +110,50 @@ export default function CreateJobModal({
   const [seed, setSeed] = React.useState(1024);
   const [numGpus, setNumGpus] = React.useState(1);
   const [ditCpuOffload, setDitCpuOffload] = React.useState(false);
+  const [ditLayerwiseOffload, setDitLayerwiseOffload] = React.useState(false);
   const [textEncoderCpuOffload, setTextEncoderCpuOffload] =
     React.useState(false);
   const [vaeCpuOffload, setVaeCpuOffload] = React.useState(false);
   const [imageEncoderCpuOffload, setImageEncoderCpuOffload] =
     React.useState(false);
   const [useFsdpInference, setUseFsdpInference] = React.useState(false);
+
+  // H3 is the only registered model with an end frame or references.
+  const supportsLastImage = modelId.toLowerCase().includes('minimax-h3');
+  const usingReferences = supportsLastImage && references.length > 0;
+
+  // JobCard re-renders on every job-list poll, so `editingJob` is a fresh
+  // object each time. Effects must depend on these, never on the object.
+  const editingJobId = editingJob?.id ?? null;
+  const editingJobModelId = editingJob?.model_id ?? null;
+
+  // Layerwise offload and FSDP compete for the DiT weights and FastVideoArgs
+  // silently picks a winner (fastvideo_args.py:859); resolve it visibly here.
+  // dit_cpu_offload is deliberately not interlocked -- it is a modifier, not a
+  // competing strategy.
+  const handleDitLayerwiseOffloadChange = React.useCallback((next: boolean) => {
+    setDitLayerwiseOffload(next);
+    if (next) {
+      setUseFsdpInference(false);
+    }
+  }, []);
+
+  const handleUseFsdpInferenceChange = React.useCallback((next: boolean) => {
+    setUseFsdpInference(next);
+    if (next) {
+      setDitLayerwiseOffload(false);
+    }
+  }, []);
+
+  const handleNumGpusChange = React.useCallback((next: number) => {
+    setNumGpus(next);
+    if (next > 1) {
+      // Dropping back to one GPU leaves FSDP alone: single-GPU FSDP is a
+      // valid way to reach its CPU offload (docs/inference/offloading.md).
+      setUseFsdpInference(true);
+      setDitLayerwiseOffload(false);
+    }
+  }, []);
   const [enableTorchCompile, setEnableTorchCompile] = React.useState(false);
   const [vsaSparsity, setVsaSparsity] = React.useState(0);
   const [tpSize, setTpSize] = React.useState(-1);
@@ -126,6 +204,47 @@ export default function CreateJobModal({
     const justOpened = isOpen && !justOpenedRef.current;
     justOpenedRef.current = isOpen;
     if (!justOpened) return;
+    if (editingJob) {
+      // Must not fall through to the defaults below: a partially-seeded
+      // form silently edits values the user never saw.
+      const f = jobToFormFields(editingJob);
+      setModelId(f.modelId);
+      setName(f.name);
+      setPrompt(f.prompt);
+      setNegativePrompt(f.negativePrompt);
+      setImagePath(f.imagePath);
+      setImageFileName(f.imagePath.split('/').pop() ?? '');
+      setLastImagePath(f.lastImagePath);
+      setLastImageFileName(f.lastImagePath.split('/').pop() ?? '');
+      setReferences(f.references);
+      setPromptFields(f.promptFields ?? EMPTY_H3_PROMPT_FIELDS);
+      setUseGuidedPrompt(f.promptFields !== null);
+      setNumInferenceSteps(f.numInferenceSteps);
+      setNumFrames(f.numFrames);
+      setHeight(f.height);
+      setWidth(f.width);
+      setGuidanceScale(f.guidanceScale);
+      setGuidanceRescale(f.guidanceRescale);
+      setFps(f.fps);
+      setSeed(f.seed);
+      setNumGpus(f.numGpus);
+      setDitCpuOffload(f.ditCpuOffload);
+      setDitLayerwiseOffload(f.ditLayerwiseOffload);
+      setTextEncoderCpuOffload(f.textEncoderCpuOffload);
+      setVaeCpuOffload(f.vaeCpuOffload);
+      setImageEncoderCpuOffload(f.imageEncoderCpuOffload);
+      setUseFsdpInference(f.useFsdpInference);
+      setEnableTorchCompile(f.enableTorchCompile);
+      setVsaSparsity(f.vsaSparsity);
+      setTpSize(f.tpSize);
+      setSpSize(f.spSize);
+      setReferenceError(null);
+      setModelLoadError(null);
+      setImageUploadError(null);
+      setLastImageUploadError(null);
+      setSubmitError(null);
+      return;
+    }
     const opts = options;
     setNumInferenceSteps(opts.numInferenceSteps);
     setNumFrames(workloadType === 't2i' ? 1 : opts.numFrames);
@@ -137,6 +256,7 @@ export default function CreateJobModal({
     setSeed(opts.seed);
     setNumGpus(opts.numGpus);
     setDitCpuOffload(opts.ditCpuOffload);
+    setDitLayerwiseOffload(opts.ditLayerwiseOffload ?? false);
     setTextEncoderCpuOffload(opts.textEncoderCpuOffload);
     setVaeCpuOffload(opts.vaeCpuOffload);
     setImageEncoderCpuOffload(opts.imageEncoderCpuOffload);
@@ -151,8 +271,16 @@ export default function CreateJobModal({
         inferenceWorkload as 't2v' | 'i2v' | 't2i',
       ),
     );
+    setName('');
     setImagePath('');
     setImageFileName('');
+    setLastImagePath('');
+    setLastImageFileName('');
+    setLastImageUploadError(null);
+    setReferences([]);
+    setReferenceError(null);
+    setPromptFields(EMPTY_H3_PROMPT_FIELDS);
+    setUseGuidedPrompt(true);
     setSelectedDatasetId('');
     setSelectedValidationDatasetId('');
     setModelLoadError(null);
@@ -168,7 +296,7 @@ export default function CreateJobModal({
       setRealScoreModelPath('');
       setFakeScoreModelPath('');
     }
-  }, [isOpen, workloadType, inferenceWorkload, options]);
+  }, [isOpen, workloadType, inferenceWorkload, options, editingJobId]);
 
   // Load the models available for this workload.
   React.useEffect(() => {
@@ -188,7 +316,16 @@ export default function CreateJobModal({
           opts,
           inferenceWorkload as 't2v' | 'i2v' | 't2i',
         );
-        const chosen = ids.includes(defaultId) ? defaultId : (list[0]?.id ?? '');
+        // When editing, the job's own model wins over the workload default --
+        // this resolves after the seeding effect, so choosing a default here
+        // would silently swap the model out from under the user.
+        const editedId = editingJobModelId;
+        const chosen =
+          editedId && ids.includes(editedId)
+            ? editedId
+            : ids.includes(defaultId)
+              ? defaultId
+              : (list[0]?.id ?? '');
         setModelId(chosen);
         if (workloadType === 'dmd_t2v') {
           setRealScoreModelPath(chosen);
@@ -210,7 +347,7 @@ export default function CreateJobModal({
     return () => {
       stale = true;
     };
-  }, [isOpen, inferenceWorkload, workloadType]);
+  }, [isOpen, inferenceWorkload, workloadType, editingJobModelId]);
 
   // Training jobs need a dataset; load the ready datasets when relevant.
   React.useEffect(() => {
@@ -262,6 +399,106 @@ export default function CreateJobModal({
     }
   }
 
+  async function handleLastImageChange(
+    e: React.ChangeEvent<HTMLInputElement>,
+  ) {
+    const file = e.target.files?.[0];
+    if (!file) {
+      setLastImagePath('');
+      setLastImageFileName('');
+      setLastImageUploadError(null);
+      return;
+    }
+    setIsUploadingLastImage(true);
+    setLastImageFileName(file.name);
+    setLastImageUploadError(null);
+    try {
+      const { path } = await uploadImage(file);
+      setLastImagePath(path);
+    } catch (error) {
+      console.error('Failed to upload end image:', error);
+      setLastImagePath('');
+      setLastImageFileName('');
+      setLastImageUploadError(
+        error instanceof Error
+          ? `${error.message}. Choose the image again to retry.`
+          : 'The image could not be uploaded. Choose it again to retry.',
+      );
+    } finally {
+      setIsUploadingLastImage(false);
+    }
+  }
+
+  async function handleAddReference(
+    e: React.ChangeEvent<HTMLInputElement>,
+  ) {
+    const file = e.target.files?.[0];
+    e.target.value = '';               // allow re-picking the same file
+    if (!file) return;
+    setIsUploadingReference(true);
+    setReferenceError(null);
+    try {
+      const { path, media_type } = await uploadMedia(file);
+      const next: H3Reference[] = [
+        ...references,
+        {
+          id: `${Date.now()}-${file.name}`,
+          source: path,
+          media_type,
+          fileName: file.name,
+        },
+      ];
+      setReferences(next);
+      setReferenceError(validateReferences(next));
+    } catch (error) {
+      console.error('Failed to upload reference:', error);
+      setReferenceError(
+        error instanceof Error ? error.message : 'The file could not be uploaded.',
+      );
+    } finally {
+      setIsUploadingReference(false);
+    }
+  }
+
+  function removeReference(id: string) {
+    const next = references.filter((r) => r.id !== id);
+    setReferences(next);
+    setReferenceError(validateReferences(next));
+  }
+
+  function seedPromptFields() {
+    setPromptFields({
+      ...EMPTY_H3_PROMPT_FIELDS,
+      ...referencePromptSeed(references),
+    });
+    setUseGuidedPrompt(true);
+  }
+
+  function setPromptField(section: string, value: string) {
+    setPromptFields((prev) => ({ ...prev, [section]: value }));
+  }
+
+  // Switching between the guided fields and the raw editor keeps whatever was
+  // typed: serialize on the way out, parse back on the way in.
+  function toggleGuidedPrompt() {
+    if (useGuidedPrompt) {
+      if (!isEmptyPromptFields(promptFields)) {
+        setPrompt(serializeH3Prompt(promptFields));
+      }
+      setUseGuidedPrompt(false);
+    } else {
+      const parsed = parseH3Prompt(prompt);
+      if (parsed) setPromptFields(parsed);
+      setUseGuidedPrompt(true);
+    }
+  }
+
+  function clearLastImage() {
+    setLastImagePath('');
+    setLastImageFileName('');
+    setLastImageUploadError(null);
+  }
+
   function clearImage() {
     setImagePath('');
     setImageFileName('');
@@ -271,7 +508,16 @@ export default function CreateJobModal({
 
   async function handleSubmit(e: React.FormEvent<HTMLFormElement>) {
     e.preventDefault();
-    if (isInference && workloadType === 'i2v' && !imagePath) return;
+    if (isInference && workloadType === 'i2v' && !imagePath && !usingReferences)
+      return;
+    if (usingReferences && validateReferences(references)) return;
+    if (
+      usingReferences &&
+      useGuidedPrompt &&
+      isEmptyPromptFields(promptFields) &&
+      !prompt.trim()
+    )
+      return;
     // Send the dataset id; the backend resolves it to the on-disk media dir.
     const effectiveDataPath = selectedDatasetId ?? '';
     if (!isInference && !selectedDatasetId) return;
@@ -285,14 +531,35 @@ export default function CreateJobModal({
     try {
       const payload: CreateJobRequest = {
         model_id: modelId,
-        prompt,
+        name: name.trim(),
+        prompt:
+          usingReferences && useGuidedPrompt && !isEmptyPromptFields(promptFields)
+            ? serializeH3Prompt(promptFields)
+            : prompt,
         workload_type: workloadType,
         job_type: effectiveJobType,
         ...(isInference
           ? {
-              ...(workloadType === 'i2v' && imagePath
+              // Ref2VA and the FL2VA keyframes are mutually exclusive:
+              // _prepare_ref2va rejects image_path/last_image_path outright
+              // when references are present.
+              ...(workloadType === 'i2v' && !usingReferences && imagePath
                 ? { image_path: imagePath }
                 : {}),
+              ...(workloadType === 'i2v' &&
+              supportsLastImage &&
+              !usingReferences &&
+              lastImagePath
+                ? { last_image_path: lastImagePath }
+                : {}),
+              ...(workloadType === 'i2v' && supportsLastImage && references.length
+                ? {
+                    references: references.map((r) => ({
+                      source: r.source,
+                      media_type: r.media_type,
+                    })),
+                  }
+                : {}),
               negative_prompt: negativePrompt,
               num_inference_steps: numInferenceSteps,
               num_frames: numFrames,
@@ -304,6 +571,7 @@ export default function CreateJobModal({
               seed,
               num_gpus: numGpus,
               dit_cpu_offload: ditCpuOffload,
+              dit_layerwise_offload: ditLayerwiseOffload,
               text_encoder_cpu_offload: textEncoderCpuOffload,
               vae_cpu_offload: vaeCpuOffload,
               image_encoder_cpu_offload: imageEncoderCpuOffload,
@@ -334,7 +602,14 @@ export default function CreateJobModal({
                 : {}),
             }),
       };
-      await createJob(payload);
+      if (editingJob) {
+        await updateJob(
+          editingJob.id,
+          payload as unknown as Record<string, unknown>,
+        );
+      } else {
+        await createJob(payload);
+      }
       onSuccess();
       onClose();
     } catch (err) {
@@ -356,9 +631,9 @@ export default function CreateJobModal({
 
   const workloadLabel =
     WORKLOAD_OPTIONS[jobType]?.find((o) => o.type === workloadType)?.label ?? '';
-  const title = `New ${jobType.charAt(0).toUpperCase() + jobType.slice(1)} Job${
-    workloadLabel ? ` (${workloadLabel})` : ''
-  }`;
+  const title = `${readOnly ? 'View' : editingJob ? 'Edit' : 'New'} ${
+    jobType.charAt(0).toUpperCase() + jobType.slice(1)
+  } Job${workloadLabel ? ` (${workloadLabel})` : ''}`;
 
   return (
     <Dialog
@@ -385,6 +660,23 @@ export default function CreateJobModal({
           autoComplete="off"
           className="flex flex-col gap-3.5"
         >
+          {/* disabled cascades to every control inside; display:contents
+              keeps the parent's flex layout. */}
+          <fieldset
+            disabled={readOnly}
+            style={{ display: 'contents' }}
+            className="contents"
+          >
+          <FieldRow htmlFor="modal-name" label="Name (optional)">
+            <Input
+              id="modal-name"
+              value={name}
+              onChange={(e) => setName(e.target.value)}
+              placeholder="Shown on the job card and used for the output filename"
+              disabled={isSubmitting}
+            />
+          </FieldRow>
+
           <FieldRow htmlFor="modal-modelId" label="Model">
             <NativeSelect
               id="modal-modelId"
@@ -429,12 +721,12 @@ export default function CreateJobModal({
                 type="file"
                 accept=".png,.jpg,.jpeg,.webp,.bmp"
                 onChange={handleImageChange}
-                disabled={isSubmitting || isUploadingImage}
+                disabled={isSubmitting || isUploadingImage || usingReferences}
                 aria-describedby={
                   imageUploadError ? 'modal-image-error' : undefined
                 }
                 aria-invalid={imageUploadError ? true : undefined}
-                required
+                required={!usingReferences}
                 className="h-auto py-2 file:mr-3 file:cursor-pointer file:rounded-md file:border-0 file:bg-secondary file:px-2 file:py-1 file:text-sm file:text-secondary-foreground"
               />
               {imageFileName && (
@@ -462,24 +754,169 @@ export default function CreateJobModal({
             </FieldRow>
           )}
 
-          <FieldRow
-            htmlFor="modal-prompt"
-            label={isInference ? 'Prompt' : 'Description'}
-          >
-            <Textarea
-              id="modal-prompt"
-              value={prompt}
-              onChange={(e) => setPrompt(e.target.value)}
-              rows={isInference ? 3 : 2}
-              placeholder={
-                isInference
-                  ? 'A curious raccoon peers through a vibrant field of yellow sunflowers…'
-                  : 'Brief description of this training job…'
-              }
-              required
-              disabled={isSubmitting}
-            />
-          </FieldRow>
+          {isInference && workloadType === 'i2v' && supportsLastImage && (
+            <FieldRow htmlFor="modal-last-image" label="End Frame (optional)">
+              <Input
+                id="modal-last-image"
+                type="file"
+                accept=".png,.jpg,.jpeg,.webp,.bmp"
+                onChange={handleLastImageChange}
+                disabled={isSubmitting || isUploadingLastImage}
+                aria-describedby={
+                  lastImageUploadError ? 'modal-last-image-error' : undefined
+                }
+                aria-invalid={lastImageUploadError ? true : undefined}
+                className="h-auto py-2 file:mr-3 file:cursor-pointer file:rounded-md file:border-0 file:bg-secondary file:px-2 file:py-1 file:text-sm file:text-secondary-foreground"
+              />
+              {lastImageFileName && (
+                <span className="mt-0.5 text-xs text-muted-foreground">
+                  {isUploadingLastImage ? 'Uploading…' : lastImageFileName} ·{' '}
+                  <button
+                    type="button"
+                    onClick={clearLastImage}
+                    disabled={isSubmitting || isUploadingLastImage}
+                    className="text-accent-blue underline-offset-2 hover:underline disabled:cursor-not-allowed disabled:opacity-50"
+                  >
+                    Clear
+                  </button>
+                </span>
+              )}
+              {lastImageUploadError && (
+                <p
+                  id="modal-last-image-error"
+                  role="alert"
+                  className="text-sm text-destructive"
+                >
+                  {lastImageUploadError}
+                </p>
+              )}
+            </FieldRow>
+          )}
+
+          {isInference && workloadType === 'i2v' && supportsLastImage && (
+            <FieldRow htmlFor="modal-reference" label="References (Ref2VA)">
+              <Input
+                id="modal-reference"
+                type="file"
+                accept=".png,.jpg,.jpeg,.webp,.bmp,.mp4,.mov,.mkv,.webm,.avi,.wav,.mp3,.flac,.m4a,.ogg"
+                onChange={handleAddReference}
+                disabled={
+                  isSubmitting ||
+                  isUploadingReference ||
+                  references.length >= H3_MAX_REFERENCES
+                }
+                className="h-auto py-2 file:mr-3 file:cursor-pointer file:rounded-md file:border-0 file:bg-secondary file:px-2 file:py-1 file:text-sm file:text-secondary-foreground"
+              />
+              {isUploadingReference && (
+                <span className="mt-0.5 text-xs text-muted-foreground">
+                  Uploading…
+                </span>
+              )}
+              {references.length > 0 && (
+                <ul className="mt-1 flex list-none flex-col gap-1 p-0">
+                  {references.map((reference, index) => (
+                    <li
+                      key={reference.id}
+                      className="flex items-center gap-2 text-xs text-muted-foreground"
+                    >
+                      <code className="font-mono text-accent-blue">
+                        {labelReferences(references)[index]}
+                      </code>
+                      <span className="truncate">{reference.fileName}</span>
+                      <button
+                        type="button"
+                        onClick={() => removeReference(reference.id)}
+                        disabled={isSubmitting}
+                        className="ml-auto text-accent-blue underline-offset-2 hover:underline disabled:cursor-not-allowed disabled:opacity-50"
+                      >
+                        Remove
+                      </button>
+                    </li>
+                  ))}
+                </ul>
+              )}
+              {references.length > 0 && (
+                <button
+                  type="button"
+                  onClick={seedPromptFields}
+                  disabled={isSubmitting}
+                  className="mt-1 self-start text-xs text-accent-blue underline-offset-2 hover:underline disabled:cursor-not-allowed disabled:opacity-50"
+                >
+                  Fill prompt sections from references
+                </button>
+              )}
+              {referenceError && (
+                <p role="alert" className="mt-0.5 text-xs text-destructive">
+                  {referenceError}
+                </p>
+              )}
+              <span className="mt-0.5 text-xs text-muted-foreground">
+                Ref2VA replaces the keyframes: up to 9 images, 3 videos, 3 audio
+                (12 total). Audio needs at least one image or video.
+              </span>
+            </FieldRow>
+          )}
+
+          {usingReferences && useGuidedPrompt ? (
+            /* Six-section format from the model's reference prompt guide. */
+            <>
+              {H3_PROMPT_SECTIONS.map((section) => (
+                <FieldRow
+                  key={section}
+                  htmlFor={`modal-prompt-${section}`}
+                  label={H3_SECTION_LABELS[section]}
+                >
+                  <Textarea
+                    id={`modal-prompt-${section}`}
+                    value={promptFields[section]}
+                    onChange={(e) => setPromptField(section, e.target.value)}
+                    rows={section === 'detailed_description' ? 5 : 2}
+                    placeholder={H3_SECTION_HINTS[section]}
+                    disabled={isSubmitting}
+                  />
+                </FieldRow>
+              ))}
+              <button
+                type="button"
+                onClick={toggleGuidedPrompt}
+                disabled={isSubmitting}
+                className="self-start text-xs text-accent-blue underline-offset-2 hover:underline disabled:cursor-not-allowed disabled:opacity-50"
+              >
+                Edit as raw prompt
+              </button>
+            </>
+          ) : (
+            <>
+              <FieldRow
+                htmlFor="modal-prompt"
+                label={isInference ? 'Prompt' : 'Description'}
+              >
+                <Textarea
+                  id="modal-prompt"
+                  value={prompt}
+                  onChange={(e) => setPrompt(e.target.value)}
+                  rows={isInference ? 3 : 2}
+                  placeholder={
+                    isInference
+                      ? 'A curious raccoon peers through a vibrant field of yellow sunflowers…'
+                      : 'Brief description of this training job…'
+                  }
+                  required={!(usingReferences && useGuidedPrompt)}
+                  disabled={isSubmitting}
+                />
+              </FieldRow>
+              {usingReferences && (
+                <button
+                  type="button"
+                  onClick={toggleGuidedPrompt}
+                  disabled={isSubmitting}
+                  className="self-start text-xs text-accent-blue underline-offset-2 hover:underline disabled:cursor-not-allowed disabled:opacity-50"
+                >
+                  Edit as prompt sections
+                </button>
+              )}
+            </>
+          )}
 
           {isInference && (
             <FieldRow htmlFor="modal-negative-prompt" label="Negative Prompt">
@@ -849,6 +1286,13 @@ export default function CreateJobModal({
                   onChange={setDitCpuOffload}
                   disabled={isSubmitting}
                 />
+                <ToggleRow
+                  id="modal-dit-layerwise-offload"
+                  label="DiT Layerwise Offload"
+                  checked={ditLayerwiseOffload}
+                  onChange={handleDitLayerwiseOffloadChange}
+                  disabled={isSubmitting}
+                />
                 <ToggleRow
                   id="modal-text-encoder-cpu-offload"
                   label="Text Encoder CPU Offload"
@@ -860,7 +1304,7 @@ export default function CreateJobModal({
                   id="modal-use-fsdp-inference"
                   label="Use FSDP Inference"
                   checked={useFsdpInference}
-                  onChange={setUseFsdpInference}
+                  onChange={handleUseFsdpInferenceChange}
                   disabled={isSubmitting}
                 />
                 <ToggleRow
@@ -891,7 +1335,7 @@ export default function CreateJobModal({
                   max={8}
                   step={1}
                   value={numGpus}
-                  onChange={setNumGpus}
+                  onChange={handleNumGpusChange}
                   disabled={isSubmitting}
                 />
                 <NumberRow
@@ -906,6 +1350,8 @@ export default function CreateJobModal({
             </details>
           )}
 
+          </fieldset>
+
           <div className="flex flex-col items-start gap-2">
             {submitError && (
               <p role="alert" className="text-sm text-destructive">
@@ -914,14 +1360,22 @@ export default function CreateJobModal({
             )}
             <Button
               type="submit"
+              hidden={readOnly}
               disabled={
+                readOnly ||
                 isSubmitting ||
                 isUploadingImage ||
                 !!modelLoadError ||
                 !!datasetLoadError
               }
             >
-              {isSubmitting ? 'Creating…' : 'Create Job'}
+              {isSubmitting
+                ? editingJob
+                  ? 'Saving…'
+                  : 'Creating…'
+                : editingJob
+                  ? 'Save Changes'
+                  : 'Create Job'}
             </Button>
           </div>
         </form>

apps/fastvideo_studio/src/components/jobs/JobCard.tsx

@@ -3,11 +3,13 @@
 import * as React from 'react';
 import { Timer } from 'lucide-react';
 
+import CreateJobModal from '@/components/jobs/CreateJobModal';
 import { Badge, type BadgeProps } from '@/components/ui/badge';
 import { Button } from '@/components/ui/button';
 import { useStore } from '@/hooks/useStore';
 import {
   deleteJob,
+  duplicateJob,
   downloadJobVideo,
   startJob,
   stopJob,
@@ -62,6 +64,8 @@ export default function JobCard({ job, onJobUpdated }: JobCardProps) {
   const isSelected = activeJobId === job.id;
 
   const [isLoading, setIsLoading] = React.useState(false);
+  const [isEditing, setIsEditing] = React.useState(false);
+  const [isViewing, setIsViewing] = React.useState(false);
   const [currentTime, setCurrentTime] = React.useState(() => Date.now());
 
   const elapsedTime = computeElapsed(job, currentTime);
@@ -103,6 +107,21 @@ export default function JobCard({ job, onJobUpdated }: JobCardProps) {
     }
   }
 
+  async function handleDuplicate(e: React.MouseEvent) {
+    e.preventDefault();
+    e.stopPropagation();
+    if (isLoading) return;
+    setIsLoading(true);
+    try {
+      await duplicateJob(job.id);
+      onJobUpdated?.();
+    } catch (err) {
+      alert(err instanceof Error ? err.message : 'Failed to duplicate job');
+    } finally {
+      setIsLoading(false);
+    }
+  }
+
   async function handleDelete(e: React.MouseEvent) {
     e.preventDefault();
     e.stopPropagation();
@@ -156,16 +175,23 @@ export default function JobCard({ job, onJobUpdated }: JobCardProps) {
       >
         <span className="flex flex-wrap items-center justify-between gap-2">
           <span className="text-[0.95rem] font-semibold text-foreground">
-            {job.model_id}
+            {job.name?.trim() || job.model_id}
           </span>
           <Badge variant={BADGE_VARIANTS[job.status] ?? 'secondary'}>
             {job.status}
           </Badge>
         </span>
         <span className="max-w-full overflow-hidden text-ellipsis whitespace-nowrap text-sm text-muted-foreground">
-          {job.prompt}
+          {job.name?.trim() ? `${job.model_id} · ${job.prompt}` : job.prompt}
         </span>
         <span className="flex flex-wrap items-center gap-4 text-xs text-muted-foreground">
+          {/* Short job id; logs and output dirs are keyed on the full UUID. */}
+          <span
+            className="font-mono text-muted-foreground/80"
+            title={job.id}
+          >
+            {job.id.slice(0, 8)}
+          </span>
           {job.job_type === 'inference' ? (
             <>
               <span>{job.num_frames} frames</span>
@@ -226,6 +252,51 @@ export default function JobCard({ job, onJobUpdated }: JobCardProps) {
               Download Video
             </Button>
           )}
+        {!(
+          job.status === 'pending' ||
+          job.status === 'failed' ||
+          job.status === 'stopped'
+        ) && (
+          <Button
+            size="sm"
+            variant="outline"
+            onClick={(e) => {
+              e.preventDefault();
+              e.stopPropagation();
+              setIsViewing(true);
+            }}
+            disabled={isLoading}
+            title="View this job's configuration"
+          >
+            View
+          </Button>
+        )}
+        {(job.status === 'pending' ||
+          job.status === 'failed' ||
+          job.status === 'stopped') && (
+          <Button
+            size="sm"
+            variant="outline"
+            onClick={(e) => {
+              e.preventDefault();
+              e.stopPropagation();
+              setIsEditing(true);
+            }}
+            disabled={isLoading}
+            title="Edit this job's configuration"
+          >
+            Edit
+          </Button>
+        )}
+        <Button
+          size="sm"
+          variant="outline"
+          onClick={handleDuplicate}
+          disabled={isLoading}
+          title="Create a new pending job with this configuration"
+        >
+          Duplicate
+        </Button>
         <Button
           size="sm"
           variant="destructive"
@@ -235,6 +306,30 @@ export default function JobCard({ job, onJobUpdated }: JobCardProps) {
           Delete
         </Button>
       </div>
+      {isViewing && (
+        <CreateJobModal
+          isOpen
+          readOnly
+          editingJob={job}
+          jobType={(job.job_type ?? 'inference') as never}
+          workloadType={job.workload_type ?? 't2v'}
+          onClose={() => setIsViewing(false)}
+          onSuccess={() => setIsViewing(false)}
+        />
+      )}
+      {isEditing && (
+        <CreateJobModal
+          isOpen
+          editingJob={job}
+          jobType={(job.job_type ?? 'inference') as never}
+          workloadType={job.workload_type ?? 't2v'}
+          onClose={() => setIsEditing(false)}
+          onSuccess={() => {
+            setIsEditing(false);
+            onJobUpdated?.();
+          }}
+        />
+      )}
     </article>
   );
 }

apps/fastvideo_studio/src/components/jobs/JobDetailsSidebar.tsx

@@ -153,9 +153,16 @@ export default function JobDetailsSidebar({
       }}
     >
       <div className="flex items-center justify-between border-b border-border px-5 py-4">
-        <h2 className="m-0 text-base font-semibold text-foreground">
-          Job Details
-        </h2>
+        <div className="min-w-0">
+          <h2 className="m-0 text-base font-semibold text-foreground">
+            Job Details
+          </h2>
+          {/* Full job id: keys ~/h3_studio_logs/<id>.log, the output directory
+              and every API route, so make it selectable for copy/paste. */}
+          <code className="mt-0.5 block select-all truncate font-mono text-xs text-muted-foreground">
+            {job.id}
+          </code>
+        </div>
         <div className="flex items-center gap-2">
           <Button
             type="button"

apps/fastvideo_studio/src/lib/api.ts

@@ -56,6 +56,8 @@ export function getJobVideoUrl(jobId: string): string {
 
 export interface CreateJobRequest {
 	model_id: string;
+	/** Optional label; the card and output filename fall back to the prompt. */
+	name?: string;
 	prompt: string;
 	workload_type?: string;
 	job_type?: JobType;
@@ -152,6 +154,65 @@ export async function updateSettings(
 	return response.json();
 }
 
+export type MediaType = "image" | "video" | "audio";
+
+/**
+ * Upload an image, video or audio file for a MiniMax-H3 Ref2VA reference.
+ * The server derives media_type from the extension and returns it, so callers
+ * do not have to duplicate that mapping.
+ */
+/** Create a new pending job with the same configuration as an existing one. */
+export async function duplicateJob(jobId: string): Promise<{ id: string }> {
+	const baseApiUrl = getApiBaseUrl();
+	const response = await fetch(`${baseApiUrl}/jobs/${jobId}/duplicate`, {
+		method: "POST",
+	});
+	if (!response.ok) {
+		const err = await response
+			.json()
+			.catch(() => ({ detail: "Duplicate failed" }));
+		throw new Error(err.detail || "Duplicate failed");
+	}
+	return response.json();
+}
+
+/** Edit a pending job's configuration. Started jobs are rejected by the API. */
+export async function updateJob(
+	jobId: string,
+	updates: Record<string, unknown>,
+): Promise<unknown> {
+	const baseApiUrl = getApiBaseUrl();
+	const response = await fetch(`${baseApiUrl}/jobs/${jobId}`, {
+		method: "PATCH",
+		headers: { "Content-Type": "application/json" },
+		body: JSON.stringify(updates),
+	});
+	if (!response.ok) {
+		const err = await response.json().catch(() => ({ detail: "Update failed" }));
+		throw new Error(err.detail || "Update failed");
+	}
+	return response.json();
+}
+
+export async function uploadMedia(
+	file: File,
+): Promise<{ path: string; media_type: MediaType }> {
+	const baseApiUrl = getApiBaseUrl();
+	const formData = new FormData();
+	formData.append("file", file);
+	const response = await fetch(`${baseApiUrl}/upload-media`, {
+		method: "POST",
+		body: formData,
+	});
+	if (!response.ok) {
+		const err = await response
+			.json()
+			.catch(() => ({ detail: "Upload failed" }));
+		throw new Error(err.detail || "Upload failed");
+	}
+	return response.json();
+}
+
 export async function uploadImage(file: File): Promise<{ path: string }> {
 	const baseApiUrl = getApiBaseUrl();
 	const formData = new FormData();

apps/fastvideo_studio/src/lib/defaultOptions.ts

@@ -16,6 +16,7 @@ export interface DefaultOptions {
 	seed: number;
 	numGpus: number;
 	ditCpuOffload: boolean;
+	ditLayerwiseOffload: boolean;
 	textEncoderCpuOffload: boolean;
 	vaeCpuOffload: boolean;
 	imageEncoderCpuOffload: boolean;
@@ -44,6 +45,7 @@ export const DEFAULT_OPTIONS: DefaultOptions = {
 	seed: 1024,
 	numGpus: 1,
 	ditCpuOffload: false,
+	ditLayerwiseOffload: false,
 	textEncoderCpuOffload: false,
 	vaeCpuOffload: false,
 	imageEncoderCpuOffload: false,

apps/fastvideo_studio/src/lib/h3Prompt.test.ts

@@ -0,0 +1,65 @@
+import { describe, expect, it } from "vitest";
+import {
+	EMPTY_H3_PROMPT_FIELDS,
+	H3_PROMPT_SECTIONS,
+	isEmptyPromptFields,
+	parseH3Prompt,
+	serializeH3Prompt,
+	type H3PromptFields,
+} from "@/lib/h3Prompt";
+
+const filled: H3PromptFields = {
+	subject_definitions: "<Subject 1> is the dog in <Picture 1>.",
+	summary: "[reference generation] The target video shows <Subject 1>.",
+	retention_analysis: "<Subject 1> (appears in [Shot 1]): fully_preserved - fur retained.",
+	detailed_description: "[Shot 1] A medium shot establishes <Subject 1>.",
+	overall_soundscape: "Room tone throughout.",
+	non_diegetic_music: "N/A",
+};
+
+describe("serializeH3Prompt", () => {
+	it("emits sections in order, flush left, blank-line separated", () => {
+		const out = serializeH3Prompt(filled);
+		expect(out).toBe(
+			[
+				"subject_definitions:\n<Subject 1> is the dog in <Picture 1>.",
+				"summary:\n[reference generation] The target video shows <Subject 1>.",
+				"retention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - fur retained.",
+				"detailed_description:\n[Shot 1] A medium shot establishes <Subject 1>.",
+				"overall_soundscape:\nRoom tone throughout.",
+				"non_diegetic_music:\nN/A",
+			].join("\n\n"),
+		);
+		// content is never indented
+		expect(out).not.toMatch(/\n {2}\S/);
+	});
+
+	it("fills blank sections with N/A rather than dropping them", () => {
+		const out = serializeH3Prompt({ ...EMPTY_H3_PROMPT_FIELDS, summary: "x" });
+		for (const s of H3_PROMPT_SECTIONS) expect(out).toContain(`${s}:`);
+		expect(out).toContain("non_diegetic_music:\nN/A");
+	});
+});
+
+describe("parseH3Prompt", () => {
+	it("round-trips a serialized prompt", () => {
+		expect(parseH3Prompt(serializeH3Prompt(filled))).toEqual(filled);
+	});
+
+	it("keeps multi-line section bodies", () => {
+		const p = parseH3Prompt("summary:\nline one\nline two\n\ndetailed_description:\nd");
+		expect(p?.summary).toBe("line one\nline two");
+		expect(p?.detailed_description).toBe("d");
+	});
+
+	it("returns null for a plain prompt", () => {
+		expect(parseH3Prompt("a toy car drives into a plush dog")).toBeNull();
+	});
+});
+
+describe("isEmptyPromptFields", () => {
+	it("detects blank vs filled", () => {
+		expect(isEmptyPromptFields(EMPTY_H3_PROMPT_FIELDS)).toBe(true);
+		expect(isEmptyPromptFields(filled)).toBe(false);
+	});
+});

apps/fastvideo_studio/src/lib/h3Prompt.ts

@@ -0,0 +1,94 @@
+/**
+ * MiniMax-H3 full-reference prompt sections. Serialization follows the worked
+ * example in the model's VIDEO_PROMPT_WRITING_GUIDE_ref_en.md: `section_name:`
+ * on its own line, content flush left, one blank line between sections.
+ */
+
+export const H3_PROMPT_SECTIONS = [
+  'subject_definitions',
+  'summary',
+  'retention_analysis',
+  'detailed_description',
+  'overall_soundscape',
+  'non_diegetic_music',
+] as const;
+
+export type H3PromptSection = (typeof H3_PROMPT_SECTIONS)[number];
+
+export type H3PromptFields = Record<H3PromptSection, string>;
+
+export const EMPTY_H3_PROMPT_FIELDS: H3PromptFields = {
+  subject_definitions: '',
+  summary: '',
+  retention_analysis: '',
+  detailed_description: '',
+  overall_soundscape: '',
+  non_diegetic_music: '',
+};
+
+export const H3_SECTION_LABELS: Record<H3PromptSection, string> = {
+  subject_definitions: 'Subject definitions',
+  summary: 'Summary',
+  retention_analysis: 'Retention analysis',
+  detailed_description: 'Detailed description',
+  overall_soundscape: 'Overall soundscape',
+  non_diegetic_music: 'Non-diegetic music',
+};
+
+/** Per-section guidance, condensed from the guide's rules for each section. */
+export const H3_SECTION_HINTS: Record<H3PromptSection, string> = {
+  subject_definitions:
+    'One line per <Subject N>. A subject may draw on several references, e.g. "<Subject 2> is the dog in <Picture 2>, <Picture 3>, and <Picture 4>."',
+  summary:
+    'One paragraph, starting with a [task type] prefix: reference generation, keyframe completion, video editing, video continuation, audio reuse, audio reference. Combine with " + ".',
+  retention_analysis:
+    'Per subject: "<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - what is retained."',
+  detailed_description:
+    'Playback order, using [Shot N] markers, timestamps, (S1) speaker tags and <d>[English] dialogue</d>.',
+  overall_soundscape: 'Ambience and physical sounds. N/A if none.',
+  non_diegetic_music:
+    'Background music audible only to the audience. N/A if none.',
+};
+
+/** True when every section is blank. */
+export function isEmptyPromptFields(fields: H3PromptFields): boolean {
+  return H3_PROMPT_SECTIONS.every((s) => !fields[s].trim());
+}
+
+/**
+ * Join the sections into the prompt string the model is given. Blank sections
+ * become "N/A" rather than being dropped, matching the guide's example.
+ */
+export function serializeH3Prompt(fields: H3PromptFields): string {
+  return H3_PROMPT_SECTIONS.map((section) => {
+    const body = fields[section].trim() || 'N/A';
+    return `${section}:\n${body}`;
+  }).join('\n\n');
+}
+
+/**
+ * Split a serialized prompt back into sections, so switching between the
+ * guided fields and the raw editor does not lose work. Returns null when the
+ * text is not in section format (a plain prompt, say).
+ */
+export function parseH3Prompt(text: string): H3PromptFields | null {
+  const fields = { ...EMPTY_H3_PROMPT_FIELDS };
+  const headings = new Set<string>(H3_PROMPT_SECTIONS);
+  let current: H3PromptSection | null = null;
+  let found = false;
+
+  for (const line of text.split('\n')) {
+    const heading = line.trim().replace(/:$/, '');
+    if (line.trim().endsWith(':') && headings.has(heading)) {
+      current = heading as H3PromptSection;
+      found = true;
+      continue;
+    }
+    if (current) fields[current] += (fields[current] ? '\n' : '') + line;
+  }
+  if (!found) return null;
+  for (const section of H3_PROMPT_SECTIONS) {
+    fields[section] = fields[section].trim();
+  }
+  return fields;
+}

apps/fastvideo_studio/src/lib/h3References.test.ts

@@ -0,0 +1,70 @@
+import { describe, expect, it } from "vitest";
+import {
+	labelReferences,
+	referencePromptSeed,
+	validateReferences,
+	type H3Reference,
+} from "@/lib/h3References";
+
+const ref = (media_type: H3Reference["media_type"], n: number): H3Reference => ({
+	id: `${media_type}-${n}`,
+	source: `/tmp/${media_type}${n}`,
+	media_type,
+	fileName: `${media_type}${n}`,
+});
+
+describe("labelReferences", () => {
+	it("numbers each media type independently, in list order", () => {
+		expect(
+			labelReferences([ref("image", 1), ref("video", 1), ref("image", 2)]),
+		).toEqual(["<Picture 1>", "<Video 1>", "<Picture 2>"]);
+	});
+});
+
+describe("validateReferences", () => {
+	it("accepts an empty list and a normal mix", () => {
+		expect(validateReferences([])).toBeNull();
+		expect(validateReferences([ref("image", 1), ref("audio", 1)])).toBeNull();
+	});
+
+	it("rejects audio-only lists", () => {
+		expect(validateReferences([ref("audio", 1)])).toMatch(/paired/);
+	});
+
+	it("enforces the per-type caps", () => {
+		const videos = [1, 2, 3, 4].map((n) => ref("video", n));
+		expect(validateReferences(videos)).toMatch(/At most 3 video/);
+	});
+
+	it("enforces the overall cap", () => {
+		const many = Array.from({ length: 13 }, (_, i) => ref("image", i));
+		expect(validateReferences(many)).toMatch(/At most 12/);
+	});
+});
+
+describe("referencePromptSeed", () => {
+	it("cites the real labels and never indents content", () => {
+		const seed = referencePromptSeed([ref("image", 1), ref("video", 1)]);
+		expect(seed.subject_definitions).toContain("<Picture 1>");
+		expect(seed.subject_definitions).toContain("<Video 1>");
+		for (const value of Object.values(seed)) {
+			expect(value).not.toMatch(/^ {2}\S/m);
+		}
+	});
+
+	it("uses the guide's retention_analysis form", () => {
+		const seed = referencePromptSeed([ref("image", 1)]);
+		expect(seed.retention_analysis).toMatch(/fully_preserved - /);
+	});
+
+	it("starts the summary with a bracketed task type", () => {
+		const seed = referencePromptSeed([ref("image", 1)]);
+		expect(seed.summary).toMatch(/^\[[a-z +]+\]/);
+	});
+
+	it("describes audio references separately", () => {
+		const seed = referencePromptSeed([ref("image", 1), ref("audio", 1)]);
+		expect(seed.subject_definitions).toContain("<Audio 1>");
+		expect(seed.retention_analysis).toContain("<Audio 1>: reference -");
+	});
+});

apps/fastvideo_studio/src/lib/h3References.ts

@@ -0,0 +1,90 @@
+/**
+ * MiniMax-H3 Ref2VA reference helpers. Labels mirror the per-type counters in
+ * `build_ref2va_presentation`, so what the user sees is what the model is shown.
+ */
+import type { MediaType } from "@/lib/api";
+
+export interface H3Reference {
+	/** stable key for React lists */
+	id: string;
+	source: string;
+	media_type: MediaType;
+	fileName: string;
+}
+
+/** Per-media-type caps enforced by validate_references (reference.py). */
+export const H3_REFERENCE_LIMITS: Record<MediaType, number> = {
+	image: 9,
+	video: 3,
+	audio: 3,
+};
+export const H3_MAX_REFERENCES = 12;
+
+const LABEL_FOR: Record<MediaType, string> = {
+	image: "Picture",
+	video: "Video",
+	audio: "Audio",
+};
+
+/** Label each reference the way the pipeline will, e.g. "<Picture 2>". */
+export function labelReferences(refs: H3Reference[]): string[] {
+	const counts: Record<MediaType, number> = { image: 0, video: 0, audio: 0 };
+	return refs.map((ref) => {
+		counts[ref.media_type] += 1;
+		return `<${LABEL_FOR[ref.media_type]} ${counts[ref.media_type]}>`;
+	});
+}
+
+/** Human-readable reason the list is invalid, or null when it is acceptable. */
+export function validateReferences(refs: H3Reference[]): string | null {
+	if (refs.length === 0) return null;
+	if (refs.length > H3_MAX_REFERENCES) {
+		return `At most ${H3_MAX_REFERENCES} references (have ${refs.length}).`;
+	}
+	const counts: Record<MediaType, number> = { image: 0, video: 0, audio: 0 };
+	for (const ref of refs) counts[ref.media_type] += 1;
+	for (const type of Object.keys(counts) as MediaType[]) {
+		if (counts[type] > H3_REFERENCE_LIMITS[type]) {
+			return `At most ${H3_REFERENCE_LIMITS[type]} ${type} references (have ${counts[type]}).`;
+		}
+	}
+	if (counts.audio === refs.length) {
+		return "Audio references must be paired with at least one image or video.";
+	}
+	return null;
+}
+
+/** Seed the guided prompt fields from the current reference list. */
+export function referencePromptSeed(
+	refs: H3Reference[],
+): Record<string, string> {
+	const labels = labelReferences(refs);
+	const visual = labels.filter((l) => !l.startsWith("<Audio"));
+	const audio = labels.filter((l) => l.startsWith("<Audio"));
+
+	const subjects = visual.map(
+		(label, i) =>
+			`<Subject ${i + 1}> is the subject from ${label}; describe appearance and distinguishing features.`,
+	);
+	for (const label of audio) {
+		subjects.push(`${label} is the audio reference; describe what it provides.`);
+	}
+
+	const retention = visual.map(
+		(_, i) =>
+			`<Subject ${i + 1}> (appears in [Shot 1]): fully_preserved - what is retained.`,
+	);
+	for (const label of audio) {
+		retention.push(`${label}: reference - how it guides the audio.`);
+	}
+
+	return {
+		subject_definitions: subjects.join("\n"),
+		summary: "[reference generation] Describe the target video and each reference's role.",
+		retention_analysis: retention.join("\n"),
+		detailed_description:
+			"[Shot 1] Describe composition, subjects, environment, lighting, action and camera movement, saying where each reference takes effect.",
+		overall_soundscape: "",
+		non_diegetic_music: "",
+	};
+}

apps/fastvideo_studio/src/lib/jobToFields.test.ts

@@ -0,0 +1,74 @@
+import { describe, expect, it } from "vitest";
+import { jobToFormFields, referenceFileName, type JobLike } from "@/lib/jobToFields";
+
+const job: JobLike = {
+	id: "abc",
+	model_id: "MiniMaxAI/MiniMax-H3",
+	prompt: "subject_definitions:\n<Subject 1> is a dog.\n\nsummary:\n[reference generation] x",
+	workload_type: "i2v",
+	references: [
+		{ source: "/uploads/9f/wukong_source.mp4", media_type: "video" },
+		{ source: "/uploads/2a/MonkeyKing_0.jpg", media_type: "image" },
+	],
+	num_frames: 141,
+	height: 768,
+	width: 1344,
+	guidance_scale: 1.0,
+	guidance_rescale: 0.1,
+	seed: 0,
+	num_gpus: 4,
+	use_fsdp_inference: true,
+};
+
+describe("jobToFormFields", () => {
+	it("carries the settings that differ from form defaults", () => {
+		const f = jobToFormFields(job);
+		expect(f.numFrames).toBe(141);
+		expect(f.height).toBe(768);
+		expect(f.width).toBe(1344);
+		expect(f.guidanceScale).toBe(1.0);
+		expect(f.guidanceRescale).toBe(0.1);
+		expect(f.seed).toBe(0);          // 0 must survive, not fall back to 1024
+		expect(f.numGpus).toBe(4);
+		expect(f.useFsdpInference).toBe(true);
+	});
+
+	it("does not let falsy-but-valid values fall through to defaults", () => {
+		const f = jobToFormFields({ ...job, seed: 0, vsa_sparsity: 0, guidance_rescale: 0 });
+		expect(f.seed).toBe(0);
+		expect(f.vsaSparsity).toBe(0);
+		expect(f.guidanceRescale).toBe(0);
+	});
+
+	it("rebuilds the reference list with readable names", () => {
+		const f = jobToFormFields(job);
+		expect(f.references).toHaveLength(2);
+		expect(f.references[0].fileName).toBe("wukong_source.mp4");
+		expect(f.references[1].media_type).toBe("image");
+		expect(new Set(f.references.map((r) => r.id)).size).toBe(2);
+	});
+
+	it("splits a six-section prompt back into fields", () => {
+		const f = jobToFormFields(job);
+		expect(f.promptFields?.subject_definitions).toBe("<Subject 1> is a dog.");
+		expect(f.promptFields?.summary).toBe("[reference generation] x");
+	});
+
+	it("returns null promptFields for a plain prompt", () => {
+		const f = jobToFormFields({ ...job, prompt: "a toy car" });
+		expect(f.promptFields).toBeNull();
+		expect(f.prompt).toBe("a toy car");
+	});
+
+	it("handles a job with no references", () => {
+		const f = jobToFormFields({ ...job, references: null });
+		expect(f.references).toEqual([]);
+	});
+});
+
+describe("referenceFileName", () => {
+	it("takes the basename", () => {
+		expect(referenceFileName("/a/b/c.mp4")).toBe("c.mp4");
+		expect(referenceFileName("c.mp4")).toBe("c.mp4");
+	});
+});

apps/fastvideo_studio/src/lib/jobToFields.ts

@@ -0,0 +1,119 @@
+/**
+ * Map a persisted job back onto the create-job form's fields. Every value the
+ * form seeds from defaults must be covered here, or edit mode silently shows
+ * defaults for whatever is missing.
+ */
+import type { H3Reference } from "@/lib/h3References";
+import { parseH3Prompt, type H3PromptFields } from "@/lib/h3Prompt";
+
+export interface JobLike {
+	id: string;
+	model_id: string;
+	name?: string;
+	prompt: string;
+	workload_type?: string;
+	job_type?: string;
+	image_path?: string;
+	last_image_path?: string;
+	references?: { source: string; media_type: string }[] | null;
+	negative_prompt?: string;
+	num_inference_steps?: number;
+	num_frames?: number;
+	height?: number;
+	width?: number;
+	guidance_scale?: number;
+	guidance_rescale?: number;
+	fps?: number;
+	seed?: number;
+	num_gpus?: number;
+	dit_cpu_offload?: boolean;
+	dit_layerwise_offload?: boolean;
+	text_encoder_cpu_offload?: boolean;
+	vae_cpu_offload?: boolean;
+	image_encoder_cpu_offload?: boolean;
+	use_fsdp_inference?: boolean;
+	enable_torch_compile?: boolean;
+	vsa_sparsity?: number;
+	tp_size?: number;
+	sp_size?: number;
+}
+
+export interface JobFormFields {
+	modelId: string;
+	name: string;
+	workloadType: string;
+	jobType: string;
+	prompt: string;
+	negativePrompt: string;
+	imagePath: string;
+	lastImagePath: string;
+	references: H3Reference[];
+	promptFields: H3PromptFields | null;
+	numInferenceSteps: number;
+	numFrames: number;
+	height: number;
+	width: number;
+	guidanceScale: number;
+	guidanceRescale: number;
+	fps: number;
+	seed: number;
+	numGpus: number;
+	ditCpuOffload: boolean;
+	ditLayerwiseOffload: boolean;
+	textEncoderCpuOffload: boolean;
+	vaeCpuOffload: boolean;
+	imageEncoderCpuOffload: boolean;
+	useFsdpInference: boolean;
+	enableTorchCompile: boolean;
+	vsaSparsity: number;
+	tpSize: number;
+	spSize: number;
+}
+
+/** Uploads keep their original basename, so this is the display name. */
+export function referenceFileName(source: string): string {
+	return source.split("/").filter(Boolean).pop() ?? source;
+}
+
+export function jobToFormFields(job: JobLike): JobFormFields {
+	const refs: H3Reference[] = (job.references ?? []).map((r, i) => ({
+		id: `${job.id}-${i}`,
+		source: r.source,
+		media_type: r.media_type as H3Reference["media_type"],
+		fileName: referenceFileName(r.source),
+	}));
+
+	return {
+		modelId: job.model_id,
+		name: job.name ?? "",
+		workloadType: job.workload_type ?? "t2v",
+		jobType: job.job_type ?? "inference",
+		prompt: job.prompt ?? "",
+		negativePrompt: job.negative_prompt ?? "",
+		imagePath: job.image_path ?? "",
+		lastImagePath: job.last_image_path ?? "",
+		references: refs,
+		// null when the prompt is not in six-section form; the caller then keeps
+		// the raw editor rather than silently dropping content into fields.
+		promptFields: parseH3Prompt(job.prompt ?? ""),
+		numInferenceSteps: job.num_inference_steps ?? 50,
+		numFrames: job.num_frames ?? 81,
+		height: job.height ?? 480,
+		width: job.width ?? 832,
+		guidanceScale: job.guidance_scale ?? 5.0,
+		guidanceRescale: job.guidance_rescale ?? 0.0,
+		fps: job.fps ?? 24,
+		seed: job.seed ?? 1024,
+		numGpus: job.num_gpus ?? 1,
+		ditCpuOffload: job.dit_cpu_offload ?? false,
+		ditLayerwiseOffload: job.dit_layerwise_offload ?? false,
+		textEncoderCpuOffload: job.text_encoder_cpu_offload ?? false,
+		vaeCpuOffload: job.vae_cpu_offload ?? false,
+		imageEncoderCpuOffload: job.image_encoder_cpu_offload ?? false,
+		useFsdpInference: job.use_fsdp_inference ?? false,
+		enableTorchCompile: job.enable_torch_compile ?? false,
+		vsaSparsity: job.vsa_sparsity ?? 0,
+		tpSize: job.tp_size ?? -1,
+		spSize: job.sp_size ?? -1,
+	};
+}

apps/fastvideo_studio/src/lib/types.ts

@@ -5,6 +5,7 @@ export type JobType = "inference" | "finetuning" | "distillation";
 export interface Job {
 	id: string;
 	model_id: string;
+	name?: string;
 	prompt: string;
 	job_type?: JobType;
 	workload_type?: string;

apps/fastvideo_studio/tests/test_duplicate_and_edit.py

@@ -0,0 +1,99 @@
+# SPDX-License-Identifier: Apache-2.0
+"""Duplicating a job's config, and editing one that has not started."""
+import uuid
+
+import pytest
+
+from fastvideo_studio.database import Database
+from fastvideo_studio.job_runner import JobRunner, JobStatus
+
+
+@pytest.fixture
+def runner(tmp_path):
+    return JobRunner(
+        output_dir=str(tmp_path / "out"),
+        log_dir=str(tmp_path / "logs"),
+        database=Database(tmp_path / "t.db"),
+    )
+
+
+def _make(runner, **over):
+    kwargs = dict(
+        job_id=str(uuid.uuid4()),
+        model_id="MiniMaxAI/MiniMax-H3",
+        prompt="p",
+        workload_type="i2v",
+        num_frames=141,
+        guidance_scale=1.0,
+        num_gpus=4,
+        references=[{"source": "/x/clip.mp4", "media_type": "video"}],
+    )
+    kwargs.update(over)
+    return runner.create_job(**kwargs)
+
+
+def test_duplicate_copies_config_but_not_runtime_state(runner):
+    src = _make(runner)
+    src.status = JobStatus.COMPLETED
+    src.output_path = "/x/out.mp4"
+
+    dup = runner.duplicate_job(src.id, str(uuid.uuid4()))
+
+    assert dup.id != src.id
+    assert dup.status is JobStatus.PENDING
+    assert dup.output_path is None
+    for field in ("model_id", "prompt", "workload_type", "num_frames",
+                  "guidance_scale", "num_gpus", "references"):
+        assert getattr(dup, field) == getattr(src, field)
+
+
+def test_duplicate_deep_copies_references(runner):
+    src = _make(runner)
+    dup = runner.duplicate_job(src.id, str(uuid.uuid4()))
+    dup.references[0]["source"] = "/changed"
+    assert src.references[0]["source"] == "/x/clip.mp4"
+
+
+def test_duplicate_unknown_job(runner):
+    with pytest.raises(ValueError, match="not found"):
+        runner.duplicate_job("nope", str(uuid.uuid4()))
+
+
+def test_edit_pending_job(runner):
+    job = _make(runner)
+    updated = runner.update_job_config(job.id, {"num_frames": 192, "seed": 7})
+    assert updated.num_frames == 192
+    assert updated.seed == 7
+
+
+@pytest.mark.parametrize("status", [JobStatus.FAILED, JobStatus.STOPPED])
+def test_edit_allows_restartable_jobs(runner, status):
+    """Editable exactly when startable: neither has produced an output."""
+    job = _make(runner)
+    job.status = status
+    assert runner.update_job_config(job.id, {"seed": 7}).seed == 7
+
+
+@pytest.mark.parametrize("status", [JobStatus.COMPLETED, JobStatus.RUNNING])
+def test_edit_rejects_jobs_with_or_producing_a_result(runner, status):
+    job = _make(runner)
+    job.status = status
+    with pytest.raises(ValueError, match="can be edited"):
+        runner.update_job_config(job.id, {"seed": 7})
+
+
+def test_edit_rejects_unknown_field(runner):
+    job = _make(runner)
+    with pytest.raises(ValueError, match="Not editable"):
+        runner.update_job_config(job.id, {"status": "completed"})
+
+
+def test_name_is_carried_by_duplicate(runner):
+    src = _make(runner, name="wukong swap v2")
+    dup = runner.duplicate_job(src.id, str(uuid.uuid4()))
+    assert dup.name == "wukong swap v2"
+
+
+def test_name_is_editable(runner):
+    job = _make(runner, name="a")
+    assert runner.update_job_config(job.id, {"name": "b"}).name == "b"

apps/fastvideo_studio/tests/test_upload_naming.py

@@ -0,0 +1,36 @@
+# SPDX-License-Identifier: Apache-2.0
+"""Uploaded files keep a readable basename under a unique directory."""
+from __future__ import annotations
+
+import pytest
+
+from fastvideo_studio.server import _safe_upload_name
+
+
+@pytest.mark.parametrize(
+    ("filename", "ext", "expected"),
+    [
+        ("wukong_source.mp4", ".mp4", "wukong_source.mp4"),
+        ("MonkeyKing_0.jpg", ".jpg", "MonkeyKing_0.jpg"),
+        ("my clip (final).mp4", ".mp4", "my_clip_final.mp4"),
+        ("../../etc/passwd.png", ".png", "passwd.png"),
+        ("/abs/path/frame.png", ".png", "frame.png"),
+        ("émoji✨.png", ".png", "moji.png"),
+        ("", ".png", "upload.png"),
+        (None, ".png", "upload.png"),
+        ("...", ".png", "upload.png"),
+    ],
+)
+def test_safe_upload_name(filename, ext, expected):
+    assert _safe_upload_name(filename, ext) == expected
+
+
+def test_long_names_are_capped():
+    out = _safe_upload_name("x" * 300 + ".png", ".png")
+    assert out == "x" * 80 + ".png"
+
+
+def test_no_path_separators_survive():
+    for bad in ("a/b.png", "a\\b.png", "../x.png"):
+        assert "/" not in _safe_upload_name(bad, ".png")
+        assert "\\" not in _safe_upload_name(bad, ".png")

docker/Dockerfile

@@ -89,11 +89,29 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
     zsh \
     vim \
     curl \
+    ffmpeg \
+    libgl1 \
+    libglib2.0-0 \
+    libx11-dev \
     gcc-11 \
     g++-11 \
     clang-11 \
+    cmake \
+    pkg-config \
+    build-essential \
+    libssl-dev \
     && rm -rf /var/lib/apt/lists/*
 
+# Rust toolchain: some dependencies only ship sdists on aarch64 and need cargo
+# to build. The dormant legacy Modal image layers the identical apt set +
+# rustup on top of this image (fastvideo/tests/modal/pr_test.py); baking both
+# here keeps its manual rollback path reproducible without changing the Slurm
+# runner's package surface.
+RUN set -o pipefail && \
+    curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --default-toolchain stable --profile minimal && \
+    /root/.cargo/bin/cargo --version && /root/.cargo/bin/rustc --version
+ENV PATH=/root/.cargo/bin:${PATH}
+
 # Set up C++20 compilers for ThunderKittens
 RUN update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 --slave /usr/bin/g++ g++ /usr/bin/g++-11
 
@@ -138,6 +156,7 @@ RUN --mount=type=cache,target=/opt/uv/cache \
     source /opt/venv/bin/activate && \
     uv pip install --upgrade pip && \
     uv pip install --excludes docker/uv-excludes ".[dev]" && \
+    python -c "import cv2; print('OpenCV', cv2.__version__)" && \
     PYTAG=cp$(echo "${PYTHON_VERSION}" | tr -d .) && \
     case "${TARGETARCH:-amd64}" in \
         amd64) \
@@ -169,26 +188,21 @@ RUN --mount=type=cache,target=/opt/uv/cache \
 # flash_attn/__init__.py), so FA2/varlen/bert_padding stay from the install above;
 # rmtree clears the wheel's stale cute files first to avoid an install conflict.
 # Then verify both survive so a broken overlay fails the build instead of shipping
-# an FA2-less image. x86 only: the FA4 stack (quack-kernels etc.) is unvalidated on
-# arm64 / GB10 (sm_121), so there we skip the overlay; FA4 is opt-in
-# (FASTVIDEO_FA4=1) and errors if set without the overlay, so leave it unset on
-# arm64 and the image runs FA3/FA2 as usual.
+# an FA2-less image. The pinned stack is validated on ARM64 GB200 (sm_100) as well
+# as x86; FA4 remains opt-in through FASTVIDEO_FA4=1 so lanes with FA2 baselines
+# keep their existing numerics.
 RUN --mount=type=cache,target=/opt/uv/cache \
     source $HOME/.local/bin/env && \
     source /opt/venv/bin/activate && \
-    if [ "${TARGETARCH}" = "arm64" ]; then \
-        echo "Skipping FA4 cute overlay on arm64 (FA4 stack unvalidated there; do not set FASTVIDEO_FA4)"; \
-    else \
-        python -c "import glob, shutil; [shutil.rmtree(d, ignore_errors=True) for d in glob.glob('/opt/venv/lib/python*/site-packages/flash_attn/cute')]" && \
-        uv pip install "flash-attn-4 @ git+https://github.com/Dao-AILab/flash-attention.git@${FA4_CUTE_REF}#subdirectory=flash_attn/cute" && \
-        python -c "import flash_attn; assert hasattr(flash_attn, 'flash_attn_func'), 'FA2 was clobbered by the cute overlay'; import flash_attn.cute; print('FA2 + FA4 cute OK')"; \
-    fi
+    python -c "import glob, shutil; [shutil.rmtree(d, ignore_errors=True) for d in glob.glob('/opt/venv/lib/python*/site-packages/flash_attn/cute')]" && \
+    uv pip install "flash-attn-4 @ git+https://github.com/Dao-AILab/flash-attention.git@${FA4_CUTE_REF}#subdirectory=flash_attn/cute" && \
+    python -c "import flash_attn; assert hasattr(flash_attn, 'flash_attn_func'), 'FA2 was clobbered by the cute overlay'; import flash_attn.cute; print('FA2 + FA4 cute OK')"
 
 COPY . .
 
 # Build immutable FastVideo kernel wheels for the published image. The requested
 # architecture remains installed for normal image users; amd64 images also carry
-# an SM89 artifact so the predominant L40S Modal lanes can reuse it exactly.
+# an SM89 artifact for L40S users and the dormant legacy rollback path.
 ARG FASTVIDEO_KERNEL_PREBUILT_DIR=/opt/fastvideo-kernel-prebuilt
 RUN --mount=type=cache,target=/opt/uv/cache \
     source $HOME/.local/bin/env && \

docs/assets/cookbook-recipes.json

@@ -1,52 +1,765 @@
 {
+  "version": 8,
   "recipes": [
     {
       "id": "fastwan21-t2v",
+      "family": "wan",
+      "stage": "inference",
       "task": "Text to video",
       "label": "FastWan2.1 1.3B (distilled + VSA)",
+      "summary": "Generate a video in three denoising steps with the distilled FastWan2.1 1.3B checkpoint and video sparse attention.",
       "model": "FastVideo/FastWan2.1-T2V-1.3B-Diffusers",
       "source": "scripts/inference/inference_wan_VSA_DMD_1_3B.yaml",
-      "command": "FASTVIDEO_ATTENTION_BACKEND=VIDEO_SPARSE_ATTN fastvideo generate --config scripts/inference/inference_wan_VSA_DMD_1_3B.yaml"
+      "command": "FASTVIDEO_ATTENTION_BACKEND=VIDEO_SPARSE_ATTN fastvideo generate --config scripts/inference/inference_wan_VSA_DMD_1_3B.yaml",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under outputs_video_dmd_1.3B/"
     },
     {
       "id": "wan22-t2v",
+      "family": "wan",
+      "stage": "inference",
       "task": "Text to video",
       "label": "Wan2.2 A14B",
+      "summary": "The maintained high-capacity Wan2.2 text-to-video example with CPU offload settings encoded in its checked-in Python source.",
       "model": "Wan-AI/Wan2.2-T2V-A14B-Diffusers",
       "source": "examples/inference/basic/basic_wan2_2.py",
-      "command": "python examples/inference/basic/basic_wan2_2.py"
+      "command": "python examples/inference/basic/basic_wan2_2.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 2, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under video_samples_wan2_2_14B_t2v/"
     },
     {
       "id": "wan21-i2v",
+      "family": "wan",
+      "stage": "inference",
       "task": "Image to video",
       "label": "Wan2.1 14B 480P",
+      "summary": "Animate an input image at 480P using the maintained Wan2.1 YAML configuration and its recorded offload settings.",
       "model": "Wan-AI/Wan2.1-I2V-14B-480P-Diffusers",
       "source": "scripts/inference/inference_wan_i2v.yaml",
-      "command": "fastvideo generate --config scripts/inference/inference_wan_i2v.yaml"
-    },
-    {
-      "id": "turbowan22-i2v",
-      "task": "Image to video",
-      "label": "TurboWan2.2 A14B",
-      "model": "loayrashid/TurboWan2.2-I2V-A14B-Diffusers",
-      "source": "examples/inference/basic/basic_turbodiffusion_i2v.py",
-      "command": "python examples/inference/basic/basic_turbodiffusion_i2v.py"
+      "command": "fastvideo generate --config scripts/inference/inference_wan_i2v.yaml",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 2, "evidence": "source-configured"},
+      "evidence": "Source-backed"
     },
     {
       "id": "wan22-ti2v",
+      "family": "wan",
+      "stage": "inference",
       "task": "Text or image to video",
       "label": "Wan2.2 TI2V 5B",
+      "summary": "Use one maintained 5B checkpoint for text-to-video or add an image input to switch the same recipe to image-to-video.",
       "model": "Wan-AI/Wan2.2-TI2V-5B-Diffusers",
       "source": "examples/inference/basic/basic_wan2_2_ti2v.py",
-      "command": "python examples/inference/basic/basic_wan2_2_ti2v.py"
+      "command": "python examples/inference/basic/basic_wan2_2_ti2v.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under video_samples_wan2_2_5B_ti2v/"
+    },
+    {
+      "id": "fastmetal-1-3b-mlx",
+      "family": "wan",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "FastMetal 1.3B",
+      "summary": "Run the released FastMetal 1.3B QAD checkpoint through FastVideo's native Apple Silicon MLX path.",
+      "model": "FastVideo/FastMetal-1.3B-QAD",
+      "source": "examples/inference/basic/mlx_wan_prompt_to_video.py",
+      "command": "hf download FastVideo/FastMetal-1.3B-QAD --local-dir ./FastMetal-1.3B-QAD\npython examples/inference/basic/mlx_wan_prompt_to_video.py --model-root ./FastMetal-1.3B-QAD --mlx-checkpoint ./FastMetal-1.3B-QAD --height 480 --width 832 --num-frames 81 --prompt \"A bird's-eye view of a misty forest valley at dawn.\" --output-path ./outputs/fastmetal_1_3b.mp4",
+      "gpu_types": ["Apple Silicon"],
+      "hardware": {
+        "platform": "mlx",
+        "accelerator": "Apple M4 Max",
+        "system_memory": "36 GB unified memory",
+        "minimum_memory": "16 GB+ unified memory",
+        "peak_memory": "3.87 GiB peak MLX memory",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1638"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "MP4 video at outputs/fastmetal_1_3b.mp4",
+      "modes": ["T2V", "temporal --fast", "spatial --fast-spatial", "two-pass --refine"],
+      "limitations": [
+        "Native MLX FastMetal T2V. Add --fast for temporal RIFE, --fast-spatial to denoise at half resolution, or --refine for a two-pass upsample. basic_mps.py is the older PyTorch MPS demo."
+      ]
+    },
+    {
+      "id": "fastmetal-5b-mlx",
+      "family": "wan",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "FastMetal 5B",
+      "summary": "Run the released Wan2.2 5B FastMetal checkpoint as MLX T2V with MLX DiT denoising and MLX TAEHV decode. The CUDA Wan2.2 TI2V 5B recipe is the image-capable path.",
+      "model": "FastVideo/FastMetal-5B-QAD",
+      "source": "examples/inference/basic/mlx_wan22_generate.py",
+      "command": "hf download FastVideo/FastMetal-5B-QAD --local-dir ./FastMetal-5B-QAD\npython examples/inference/basic/mlx_wan22_generate.py --mlx-checkpoint ./FastMetal-5B-QAD --text-encoder-root ./FastMetal-5B-QAD --vae-root ./FastMetal-5B-QAD/vae --height 704 --width 1280 --num-frames 81 --prompt \"A cinematic portrait with soft neon lighting and smooth camera motion.\" --output-path ./outputs/fastmetal_5b.mp4",
+      "gpu_types": ["Apple Silicon"],
+      "hardware": {
+        "platform": "mlx",
+        "accelerator": "Apple M4 Max",
+        "system_memory": "36 GB unified memory",
+        "minimum_memory": "16 GB+ unified memory",
+        "peak_memory": "9.34 GiB peak MLX memory",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1638"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "MP4 video at outputs/fastmetal_5b.mp4",
+      "modes": ["T2V", "temporal --fast", "spatial --fast-spatial", "two-pass --refine"],
+      "limitations": [
+        "The checked-in MLX example is T2V. Image-to-video is not in mlx_wan22_generate.py. Add --fast, --fast-spatial, or --refine on the same script."
+      ]
+    },
+    {
+      "id": "fastmetal-14b-mlx",
+      "family": "wan",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "FastMetal 14B",
+      "summary": "Run the released 14B FastMetal QAD checkpoint through the same Apple Silicon MLX entrypoint as the 1.3B release.",
+      "model": "FastVideo/FastMetal-14B-QAD",
+      "source": "examples/inference/basic/mlx_wan_prompt_to_video.py",
+      "command": "hf download FastVideo/FastMetal-14B-QAD --local-dir ./FastMetal-14B-QAD\npython examples/inference/basic/mlx_wan_prompt_to_video.py --model-root ./FastMetal-14B-QAD --mlx-checkpoint ./FastMetal-14B-QAD --height 480 --width 832 --num-frames 81 --prompt \"A wide cinematic landscape at sunrise.\" --output-path ./outputs/fastmetal_14b.mp4",
+      "gpu_types": ["Apple Silicon"],
+      "hardware": {
+        "platform": "mlx",
+        "accelerator": "Apple M4 Max",
+        "system_memory": "36 GB unified memory",
+        "minimum_memory": "36 GB+ unified memory",
+        "peak_memory": "21.68 GiB peak MLX memory",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1638"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "MP4 video at outputs/fastmetal_14b.mp4",
+      "modes": ["T2V", "temporal --fast", "spatial --fast-spatial", "two-pass --refine"],
+      "limitations": [
+        "Native MLX FastMetal T2V on 36 GB+ unified memory. Same --fast, --fast-spatial, and --refine flags as the 1.3B script."
+      ]
+    },
+    {
+      "id": "turbodiffusion-wan21-1-3b-t2v",
+      "family": "turbodiffusion",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "TurboWan2.1 1.3B",
+      "summary": "A TurboDiffusion-accelerated Wan2.1 1.3B text-to-video run from its maintained single-GPU example.",
+      "model": "loayrashid/TurboWan2.1-T2V-1.3B-Diffusers",
+      "source": "examples/inference/basic/basic_turbodiffusion.py",
+      "command": "python examples/inference/basic/basic_turbodiffusion.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under video_samples_turbodiffusion/",
+      "related": ["turbodiffusion-wan21-14b-t2v", "turbowan22-i2v"]
+    },
+    {
+      "id": "turbodiffusion-wan21-14b-t2v",
+      "family": "turbodiffusion",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "TurboWan2.1 14B",
+      "summary": "TurboDiffusion acceleration applied to the 14B Wan2.1 text-to-video checkpoint; the checked-in source is configured for two GPUs.",
+      "model": "loayrashid/TurboWan2.1-T2V-14B-Diffusers",
+      "source": "examples/inference/basic/basic_turbodiffusion_14b.py",
+      "command": "python examples/inference/basic/basic_turbodiffusion_14b.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 2, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under video_samples_turbodiffusion_14B/",
+      "related": ["turbodiffusion-wan21-1-3b-t2v", "turbowan22-i2v"]
+    },
+    {
+      "id": "turbowan22-i2v",
+      "family": "turbodiffusion",
+      "stage": "inference",
+      "task": "Image to video",
+      "label": "TurboWan2.2 A14B",
+      "summary": "A one-to-four-step image-to-video path using TurboDiffusion and the SLA attention backend from its maintained example.",
+      "model": "loayrashid/TurboWan2.2-I2V-A14B-Diffusers",
+      "source": "examples/inference/basic/basic_turbodiffusion_i2v.py",
+      "command": "python examples/inference/basic/basic_turbodiffusion_i2v.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 2, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "related": ["turbodiffusion-wan21-14b-t2v"]
+    },
+    {
+      "id": "ltx2-distilled-t2v",
+      "family": "ltx2",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "LTX-2 distilled",
+      "summary": "The distilled LTX-2 text-to-video checkpoint with audio, from its maintained example. The source is configured for four GPUs.",
+      "model": "FastVideo/LTX2-Distilled-Diffusers",
+      "source": "examples/inference/basic/basic_ltx2_distilled.py",
+      "command": "python examples/inference/basic/basic_ltx2_distilled.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 4, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 (video with audio) at outputs_video/ltx2_basic/output_ltx2_distilled_t2v.mp4",
+      "related": ["ltx23-base-t2v"]
+    },
+    {
+      "id": "ltx23-base-t2v",
+      "family": "ltx2",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "LTX-2 base (1088p)",
+      "summary": "Base LTX-2 text-to-video at 1088x1920 using FastVideo default sampling for LTX2 base. The example loads a community Diffusers mirror of the base checkpoint; registered aliases include Lightricks/LTX-2 and FastVideo/LTX2-Diffusers.",
+      "model": "Davids048/LTX2-Base-Diffusers",
+      "source": "examples/inference/basic/basic_ltx2.py",
+      "command": "python examples/inference/basic/basic_ltx2.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 (video with audio) at outputs_video/ltx2_basic/output_ltx2_base_t2v_1088_1920_1.1.mp4",
+      "limitations": ["The maintained example loads the Davids048/LTX2-Base-Diffusers community mirror rather than a Lightricks upstream ID."],
+      "related": ["ltx2-distilled-t2v"]
+    },
+    {
+      "id": "hy15-t2v-480p",
+      "family": "hunyuan",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "HunyuanVideo 1.5 480P",
+      "summary": "HunyuanVideo 1.5 text-to-video at 480P with CPU offload enabled in the checked-in source for smaller GPUs.",
+      "model": "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v",
+      "source": "examples/inference/basic/basic_hy15.py",
+      "command": "python examples/inference/basic/basic_hy15.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under video_samples_hy15/",
+      "related": ["hy15-1080p-upscale"]
+    },
+    {
+      "id": "hy15-1080p-upscale",
+      "family": "hunyuan",
+      "stage": "inference",
+      "task": "Text to video (upscaled)",
+      "label": "HunyuanVideo 1.5 1080P upscale",
+      "summary": "Run HunyuanVideo 1.5 through the 480p to 720p to 1080p upscale chain in one maintained script.",
+      "model": "weizhou03/HunyuanVideo-1.5-Diffusers-1080p-2SR",
+      "source": "examples/inference/basic/basic_hy15_1080p.py",
+      "command": "python examples/inference/basic/basic_hy15_1080p.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under video_samples_hy15_1080p/",
+      "related": ["hy15-t2v-480p"]
+    },
+    {
+      "id": "cosmos25-t2w",
+      "family": "cosmos",
+      "stage": "inference",
+      "task": "Text to world",
+      "label": "Cosmos Predict 2.5 2B",
+      "summary": "Generate a navigable world video from a text prompt with Cosmos Predict 2.5 2B on a single GPU.",
+      "model": "KyleShao/Cosmos-Predict2.5-2B-Diffusers",
+      "source": "examples/inference/basic/basic_cosmos2_5_t2w.py",
+      "command": "python examples/inference/basic/basic_cosmos2_5_t2w.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed"
+    },
+    {
+      "id": "kandinsky5-t2v-lite-sft",
+      "family": "kandinsky5",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "Kandinsky 5.0 T2V Lite SFT",
+      "summary": "Kandinsky 5.0 text-to-video (Lite SFT variant) from the maintained example; alternative Lite/Pro checkpoints are listed in the source.",
+      "model": "kandinskylab/Kandinsky-5.0-T2V-Lite-sft-5s-Diffusers",
+      "source": "examples/inference/basic/basic_kandinsky5_t2v.py",
+      "command": "python examples/inference/basic/basic_kandinsky5_t2v.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {
+        "platform": "cuda",
+        "gpu_count": 1,
+        "accelerator": "NVIDIA B200",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1471"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "MP4 videos under video_samples_kandinsky5_t2v/",
+      "related": ["kandinsky5-i2v-pro-distilled"]
+    },
+    {
+      "id": "kandinsky5-i2v-pro-distilled",
+      "family": "kandinsky5",
+      "stage": "inference",
+      "task": "Image to video",
+      "label": "Kandinsky 5.0 I2V Pro distilled",
+      "summary": "Animate an input image with Kandinsky 5.0 I2V Pro (distilled) on a single GPU.",
+      "model": "kandinskylab/Kandinsky-5.0-I2V-Pro-distilled-5s-Diffusers",
+      "source": "examples/inference/basic/basic_kandinsky5_i2v.py",
+      "command": "python examples/inference/basic/basic_kandinsky5_i2v.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {
+        "platform": "cuda",
+        "gpu_count": 1,
+        "accelerator": "NVIDIA B200",
+        "peak_memory": "10,365.89 MB peak GPU memory",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1471"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "MP4 videos under video_samples_kandinsky5_i2v/",
+      "related": ["kandinsky5-t2v-lite-sft"]
+    },
+    {
+      "id": "flux2-klein-t2i",
+      "family": "flux",
+      "stage": "inference",
+      "task": "Text to image",
+      "label": "FLUX.2 Klein 4B",
+      "summary": "Generate an image in four denoising steps with the distilled FLUX.2 Klein checkpoint.",
+      "model": "black-forest-labs/FLUX.2-klein-4B",
+      "source": "examples/inference/basic/basic_flux2_klein.py",
+      "command": "python examples/inference/basic/basic_flux2_klein.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "PNG image at outputs/flux2/flux2_klein.png",
+      "related": ["flux2-dev-t2i"]
+    },
+    {
+      "id": "flux2-dev-t2i",
+      "family": "flux",
+      "stage": "inference",
+      "task": "Text to image",
+      "label": "FLUX.2 dev",
+      "summary": "Full FLUX.2 dev text-to-image with embedded guidance and the Mistral3 text encoder, from its maintained example.",
+      "model": "black-forest-labs/FLUX.2-dev",
+      "source": "examples/inference/basic/basic_flux2.py",
+      "command": "python examples/inference/basic/basic_flux2.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "PNG image at outputs/flux2/flux2.png",
+      "related": ["flux2-klein-t2i"]
+    },
+    {
+      "id": "flux1-dev-t2i",
+      "family": "flux",
+      "stage": "inference",
+      "task": "Text to image",
+      "label": "FLUX.1 dev",
+      "summary": "FLUX.1 dev text-to-image through the Diffusers-backed pipeline. The example defaults to a local weights directory, so this recipe passes the Hugging Face ID explicitly.",
+      "model": "black-forest-labs/FLUX.1-dev",
+      "source": "examples/inference/basic/basic_flux_dev.py",
+      "command": "python examples/inference/basic/basic_flux_dev.py --model-path black-forest-labs/FLUX.1-dev",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "PNG images under outputs/flux_dev/samples/",
+      "limitations": ["FLUX.1 is loadable by ID but registers no model_family in fastvideo/registry.py; it is grouped under FLUX for documentation only."]
+    },
+    {
+      "id": "glm-image-t2i",
+      "family": "glm_image",
+      "stage": "inference",
+      "task": "Text to image",
+      "label": "GLM-Image",
+      "summary": "GLM-Image text-to-image generation from its maintained example.",
+      "model": "zai-org/GLM-Image",
+      "source": "examples/inference/basic/basic_glm_image.py",
+      "command": "python examples/inference/basic/basic_glm_image.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "PNG image at image_output/landscape.png",
+      "related": ["glm-image-edit"]
+    },
+    {
+      "id": "glm-image-edit",
+      "family": "glm_image",
+      "stage": "inference",
+      "task": "Image editing",
+      "label": "GLM-Image editing",
+      "summary": "Edit an input image with an instruction prompt using GLM-Image, from its maintained editing example.",
+      "model": "zai-org/GLM-Image",
+      "source": "examples/inference/basic/edit_glm_image.py",
+      "command": "python examples/inference/basic/edit_glm_image.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "PNG image at image_output/edited.png (input: assets/images/couple.jpg)",
+      "related": ["glm-image-t2i"]
+    },
+    {
+      "id": "zimage-turbo-t2i",
+      "family": "zimage",
+      "stage": "inference",
+      "task": "Text to image",
+      "label": "Z-Image Turbo",
+      "summary": "Z-Image Turbo text-to-image on a single GPU from its maintained example.",
+      "model": "Tongyi-MAI/Z-Image-Turbo",
+      "source": "examples/inference/basic/basic_zimage.py",
+      "command": "python examples/inference/basic/basic_zimage.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "PNG image at outputs/zimage/zimage_turbo.png"
+    },
+    {
+      "id": "sd35-medium-t2i",
+      "family": "sd35",
+      "stage": "inference",
+      "task": "Text to image",
+      "label": "Stable Diffusion 3.5 Medium",
+      "summary": "Stable Diffusion 3.5 Medium text-to-image over a small built-in prompt set, from its maintained example.",
+      "model": "stabilityai/stable-diffusion-3.5-medium",
+      "source": "examples/inference/basic/basic_sd35_t2i.py",
+      "command": "python examples/inference/basic/basic_sd35_t2i.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "PNG images under outputs/sd35/samples/"
+    },
+    {
+      "id": "minimax-h3-t2v",
+      "family": "minimax_h3",
+      "stage": "inference",
+      "task": "Text to video (with audio)",
+      "label": "MiniMax H3 T2VA",
+      "summary": "Generate synchronized video and stereo audio from a structured text prompt with the full MiniMax H3 checkpoint.",
+      "model": "MiniMaxAI/MiniMax-H3",
+      "source": "examples/inference/basic/basic_minimax_h3_t2v.py",
+      "command": "python examples/inference/basic/basic_minimax_h3_t2v.py --prompt \"(S1) A presenter says <d>[English] FastVideo runs MiniMax H3.</d>\"",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"platform": "cuda", "gpu_count": 4, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 with synchronized audio at outputs/minimax_h3_t2v/minimax_h3_t2v.mp4",
+      "modes": ["T2VA"],
+      "knobs": [
+        {"key": "num_gpus", "label": "GPUs", "hint": "Sequence-parallel degree", "flag": "--num-gpus", "options": [1, 2, 4, 8], "default": 4}
+      ],
+      "limitations": ["The checked-in example defaults to four-way sequence parallelism. It does not record a GPU model or memory requirement."]
+    },
+    {
+      "id": "fasth3-preview-cuda",
+      "group": "fasth3-preview",
+      "group_label": "FastH3 Preview",
+      "group_task": "4-step text to video + audio",
+      "family": "minimax_h3",
+      "stage": "inference",
+      "task": "Few-step text to video (with audio)",
+      "label": "FastH3 Preview on CUDA",
+      "summary": "Run the DMD2-distilled FastH3 Preview with four DiT forwards, trained H3 sparse attention, compiled decode, and synchronized audio.",
+      "model": "FastVideo/FastVideo-Minimax-FastH3-Preview-v0.2",
+      "source": "examples/inference/basic/basic_fasth3.py",
+      "serving": {
+        "source": "examples/serving/openai_fasth3.yaml",
+        "install": "UV_TORCH_BACKEND=cu130 uv pip install -e \".[fasth3]\""
+      },
+      "command": "UV_TORCH_BACKEND=cu130 uv pip install -e \".[fasth3]\"\npython examples/inference/basic/basic_fasth3.py --prompt \"(S1) A presenter says <d>[English] FastVideo runs FastH3.</d>\" --profile all",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {
+        "platform": "cuda",
+        "gpu_count": 4,
+        "accelerator": "NVIDIA GB200",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1731"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "Warmup and measured MP4 files under outputs/fasth3/",
+      "modes": ["T2VA", "4-step FastH3"],
+      "knobs": [
+        {"key": "num_gpus", "label": "GPUs", "hint": "Sequence-parallel degree", "flag": "--num-gpus", "options": [1, 2, 4, 8], "default": 4},
+        {"key": "video_decode_backend", "label": "VAE decode", "hint": "Fidelity vs. speed", "flag": "--video-decode-backend", "options": [{"value": "h3-vae", "label": "Full H3 VAE"}, {"value": "taeh3", "label": "TAEH3 preview"}], "default": "h3-vae"}
+      ],
+      "limitations": ["The default all profile is the measured GB200 performance route and can change floating-point operation order. Use --profile strict --no-inference-torch-compile for the eager strict route."]
+    },
+    {
+      "id": "fasth3-preview-mlx",
+      "serving": {
+        "source": "examples/serving/mlx_fasth3.yaml",
+        "install": "uv pip install -e \".[mlx]\"",
+        "prepare": "hf download FastVideo/FastVideo-Minimax-FastH3-Preview-v0.2 --local-dir ./FastH3-Preview-v0.2\npython scripts/checkpoint_conversion/convert_minimax_h3_mlx.py --model-root ./FastH3-Preview-v0.2/transformer --out ./FastH3-MLX --formats \"int6\""
+      },
+      "group": "fasth3-preview",
+      "group_label": "FastH3 Preview",
+      "group_task": "4-step text to video + audio",
+      "family": "minimax_h3",
+      "stage": "inference",
+      "task": "Few-step text to video (with audio)",
+      "label": "FastH3 Preview on MLX",
+      "summary": "Run FastH3 Preview on Apple Silicon with a locally converted INT6 DiT, streamed Qwen3-VL conditioning, and native MLX video and audio VAEs.",
+      "model": "FastVideo/FastVideo-Minimax-FastH3-Preview-v0.2",
+      "source": "examples/inference/basic/mlx_fasth3.py",
+      "command": "hf download FastVideo/FastVideo-Minimax-FastH3-Preview-v0.2 --local-dir ./FastH3-Preview-v0.2\npython scripts/checkpoint_conversion/convert_minimax_h3_mlx.py --model-root ./FastH3-Preview-v0.2/transformer --out ./FastH3-MLX --formats \"int6\"\npython examples/inference/basic/mlx_fasth3.py --model-root ./FastH3-Preview-v0.2 --mlx-checkpoint ./FastH3-MLX/int6 --prompt \"(S1) A presenter says <d>[English] FastVideo runs FastH3.</d>\" --height 480 --width 832 --num-frames 124 --seed 2026 --output-path ./outputs/fasth3_int6.mp4",
+      "gpu_types": ["Apple Silicon"],
+      "hardware": {
+        "platform": "mlx",
+        "accelerator": "Apple M4 Max",
+        "system_memory": "36 GB unified memory",
+        "peak_memory": "19.63 GiB peak MLX memory during denoising",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1770"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "MP4 with H.264 video and stereo AAC audio at outputs/fasth3_int6.mp4",
+      "modes": ["T2VA", "temporal --fast", "spatial --fast-spatial", "opt-in VSA"],
+      "knobs": [
+        {"key": "video_decode_backend", "label": "VAE decode", "hint": "Fidelity vs. speed", "flag": "--video-decode-backend", "options": [{"value": "h3-vae", "label": "Full H3 VAE"}, {"value": "taeh3", "label": "TAEH3 preview"}], "default": "h3-vae"}
+      ],
+      "limitations": [
+        "The MLX path supports T2VA, optional temporal --fast, optional spatial --fast-spatial, and opt-in VSA on --include-vsa checkpoints. FL2VA, Ref2VA, and two-pass refinement are not wired."
+      ]
+    },
+    {
+      "id": "fasth3-preview-spark",
+      "group": "fasth3-preview",
+      "group_label": "FastH3 Preview",
+      "group_task": "4-step text to video + audio",
+      "family": "minimax_h3",
+      "stage": "inference",
+      "task": "Few-step text to video (with audio)",
+      "label": "FastH3 Preview on one DGX Spark",
+      "summary": "Run FastH3 Preview on one GB10 with Triton VSA, FA4 off, and lazy module load. Height, width, frames, and steps in the YAML are examples.",
+      "model": "FastVideo/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree",
+      "source": "examples/inference/basic/basic_fasth3_spark.yaml",
+      "serving": {
+        "source": "examples/serving/openai_fasth3_spark.yaml",
+        "install": "UV_TORCH_BACKEND=cu130 uv pip install -e .",
+        "env": "FASTVIDEO_VSA_SM100A=0 FASTVIDEO_FA4=0 FASTVIDEO_ATTENTION_BACKEND=VIDEO_SPARSE_ATTN_H3"
+      },
+      "command": "FASTVIDEO_VSA_SM100A=0 FASTVIDEO_FA4=0 FASTVIDEO_ATTENTION_BACKEND=VIDEO_SPARSE_ATTN_H3 FASTVIDEO_STAGE_LOGGING=1 fastvideo generate --config examples/inference/basic/basic_fasth3_spark.yaml",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {
+        "platform": "cuda",
+        "device": "spark",
+        "gpu_count": 1,
+        "evidence": "source-configured"
+      },
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 under outputs/fasth3_spark/",
+      "modes": ["T2VA", "1-Spark"],
+      "limitations": [
+        "Install from the DGX Spark guide, not the generic CUDA extra. GB10 has no FA4 / sm_100a VSA kernel; keep FASTVIDEO_FA4=0 and FASTVIDEO_VSA_SM100A=0.",
+        "Legal num_frames values are 17n+5, capped at 345 (15 s). Native 16:9 sizes include 832x480 and 1344x768.",
+        "Lazy module load reloads Qwen3-VL and the DiT between phases of each request. Do not pass --no-lazy-module-load on this box.",
+        "A 345-frame request on one Spark can OOM. Prefer 124 or 243 frames, TAEH3 decode, or two Sparks over QSFP."
+      ]
+    },
+    {
+      "id": "fasth3-spark-pair",
+      "group": "fasth3-preview",
+      "group_label": "FastH3 Preview",
+      "group_task": "4-step text to video + audio",
+      "family": "minimax_h3",
+      "stage": "inference",
+      "task": "Few-step text to video (with audio)",
+      "label": "FastH3 Preview on two DGX Sparks",
+      "summary": "Run one FastH3 clip across two GB10s with Ray sequence parallel over QSFP RoCE. Sequential load and lazy module load stay on because SP replicates the DiT on each node.",
+      "model": "FastVideo/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree",
+      "source": "examples/inference/basic/basic_fasth3_spark_pair.yaml",
+      "command": "source examples/inference/optimizations/spark_pair_env.sh && FASTVIDEO_VSA_SM100A=0 FASTVIDEO_FA4=0 FASTVIDEO_ATTENTION_BACKEND=VIDEO_SPARSE_ATTN_H3 FASTVIDEO_VAE_PARALLEL_DECODE=1 fastvideo generate --config examples/inference/basic/basic_fasth3_spark_pair.yaml",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {
+        "platform": "cuda",
+        "device": "spark",
+        "gpu_count": 2,
+        "accelerator": "NVIDIA GB10 (DGX Spark pair)",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1803"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "MP4 under outputs/fasth3_spark_pair/",
+      "modes": ["T2VA", "2-Spark SP"],
+      "limitations": [
+        "Requires a two-node Ray cluster on the QSFP interconnect. There is no cookbook server for this path; use Python / generate.",
+        "Height, width, frames, and steps in the YAML are examples. Edit them or pass CLI flags. See docs/getting_started/installation/spark_pair.md."
+      ]
+    },
+    {
+      "id": "minimax-h3-fl2va",
+      "family": "minimax_h3",
+      "stage": "inference",
+      "task": "First/last frame to video (with audio)",
+      "label": "MiniMax H3 FL2VA",
+      "summary": "Animate a first frame, optionally guide the final frame, and generate synchronized audio with the full MiniMax H3 checkpoint.",
+      "model": "MiniMaxAI/MiniMax-H3",
+      "source": "examples/inference/basic/basic_minimax_h3_fl2va.py",
+      "command": "python examples/inference/basic/basic_minimax_h3_fl2va.py --image path/to/first-frame.png --prompt \"(S1) The subject turns toward the camera and says <d>[English] Hello.</d>\"",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"platform": "cuda", "gpu_count": 4, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 with synchronized audio at outputs/minimax_h3_fl2va/minimax_h3_fl2va.mp4",
+      "modes": ["FL2VA"],
+      "knobs": [
+        {"key": "num_gpus", "label": "GPUs", "hint": "Sequence-parallel degree", "flag": "--num-gpus", "options": [1, 2, 4, 8], "default": 4}
+      ],
+      "limitations": ["Pass --last-image to constrain the final frame. The checked-in source defaults to four GPUs."]
+    },
+    {
+      "id": "minimax-h3-ref2va",
+      "family": "minimax_h3",
+      "stage": "inference",
+      "task": "Reference media to video (with audio)",
+      "label": "MiniMax H3 Ref2VA",
+      "summary": "Condition H3 on an ordered reference video and optional audio reference, then generate a new synchronized video and audio result.",
+      "model": "MiniMaxAI/MiniMax-H3",
+      "source": "examples/inference/basic/basic_minimax_h3_ref2va.py",
+      "command": "python examples/inference/basic/basic_minimax_h3_ref2va.py --reference-video path/to/reference.mp4 --prompt \"Create a new scene that preserves the reference identity and motion language.\"",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"platform": "cuda", "gpu_count": 4, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 with synchronized audio at outputs/minimax_h3_ref2va/minimax_h3_ref2va.mp4",
+      "modes": ["Ref2VA"],
+      "knobs": [
+        {"key": "num_gpus", "label": "GPUs", "hint": "Sequence-parallel degree", "flag": "--num-gpus", "options": [1, 2, 4, 8], "default": 4}
+      ],
+      "limitations": ["Pass --reference-audio for an additional audio reference. The checked-in source defaults to four GPUs."]
+    },
+    {
+      "id": "fasth3-lora-preview",
+      "family": "minimax_h3",
+      "stage": "inference",
+      "task": "LoRA-adapted few-step video (with audio)",
+      "label": "FastH3 LoRA Preview",
+      "summary": "Apply a FastH3 preview adapter at load time while keeping the shared four-forward performance profile and synchronized audio output.",
+      "model": "MiniMaxAI/MiniMax-H3",
+      "source": "examples/inference/basic/basic_fasth3_lora_preview.py",
+      "command": "python examples/inference/basic/basic_fasth3_lora_preview.py --lora-path path/to/adapter.safetensors --prompt \"(S1) A presenter says <d>[English] This is an adapted Fast H3 run.</d>\"",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {
+        "platform": "cuda",
+        "gpu_count": 4,
+        "accelerator": "NVIDIA B200",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1771"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "Warmup and measured MP4 files under outputs/fasth3_lora_preview/",
+      "modes": ["T2VA", "FastH3 LoRA"],
+      "knobs": [
+        {"key": "num_gpus", "label": "GPUs", "hint": "Sequence-parallel degree", "flag": "--num-gpus", "options": [1, 2, 4, 8], "default": 4},
+        {"key": "video_decode_backend", "label": "VAE decode", "hint": "Fidelity vs. speed", "flag": "--video-decode-backend", "options": [{"value": "h3-vae", "label": "Full H3 VAE"}, {"value": "taeh3", "label": "TAEH3 preview"}], "default": "h3-vae"}
+      ],
+      "limitations": ["Supply a compatible FastH3 adapter. The script infers dense or VSA attention from the adapter payload unless you override it."]
+    },
+    {
+      "id": "longcat-t2v",
+      "family": "longcat",
+      "stage": "inference",
+      "task": "Text to video",
+      "label": "LongCat Video T2V",
+      "summary": "LongCat Video text-to-video at 480p (50 steps), with distilled and 720p refinement passes included in the same maintained script.",
+      "model": "FastVideo/LongCat-Video-T2V-Diffusers",
+      "source": "examples/inference/basic/basic_longcat_t2v.py",
+      "command": "python examples/inference/basic/basic_longcat_t2v.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under outputs_video/longcat_t2v_basic/, longcat_t2v_distill/, and longcat_t2v_refine_720p/",
+      "related": ["longcat-i2v"]
+    },
+    {
+      "id": "longcat-i2v",
+      "family": "longcat",
+      "stage": "inference",
+      "task": "Image to video",
+      "label": "LongCat Video I2V",
+      "summary": "LongCat Video image-to-video with optional distilled and refinement passes, from its maintained example.",
+      "model": "FastVideo/LongCat-Video-I2V-Diffusers",
+      "source": "examples/inference/basic/basic_longcat_i2v.py",
+      "command": "python examples/inference/basic/basic_longcat_i2v.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under outputs_video/longcat_i2v_basic/ and longcat_i2v_distill/",
+      "related": ["longcat-t2v"]
+    },
+    {
+      "id": "stable-audio-open-t2a",
+      "family": "stable_audio",
+      "stage": "inference",
+      "task": "Text to audio",
+      "label": "Stable Audio Open 1.0",
+      "summary": "Six-second text-to-audio generation with Stable Audio Open 1.0 from its maintained example; duration and steps are documented knobs in the source.",
+      "model": "FastVideo/stable-audio-open-1.0-Diffusers",
+      "source": "examples/inference/basic/basic_stable_audio.py",
+      "command": "python examples/inference/basic/basic_stable_audio.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {
+        "platform": "cuda",
+        "gpu_count": 1,
+        "accelerator": "NVIDIA B200",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1260"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "WAV audio at outputs_audio/stable_audio_basic/output_stable_audio.wav",
+      "limitations": ["Must load the FastVideo converted Diffusers repo; upstream stabilityai monolithic checkpoints are not loader-compatible (see scripts/checkpoint_conversion/stable_audio_to_diffusers.py)."],
+      "related": ["stable-audio-small-t2a"]
+    },
+    {
+      "id": "stable-audio-small-t2a",
+      "family": "stable_audio",
+      "stage": "inference",
+      "task": "Text to audio",
+      "label": "Stable Audio Open Small",
+      "summary": "The smaller Stable Audio Open variant with its own shorter training window, from its maintained example.",
+      "model": "FastVideo/stable-audio-open-small-Diffusers",
+      "source": "examples/inference/basic/basic_stable_audio_small.py",
+      "command": "python examples/inference/basic/basic_stable_audio_small.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "related": ["stable-audio-open-t2a"]
+    },
+    {
+      "id": "mmaudio-v2a",
+      "family": "mmaudio",
+      "stage": "inference",
+      "task": "Video/Text to audio",
+      "label": "MMAudio large 44k v2",
+      "summary": "Add synchronized audio to a video (or from a prompt) with MMAudio large 44k v2. The example reads the model path from MMAUDIO_MODEL_PATH; this recipe passes the converted Hugging Face repo explicitly.",
+      "model": "FastVideo/MMAudio-large-44k-v2-Diffusers",
+      "source": "examples/inference/basic/basic_mmaudio.py",
+      "command": "MMAUDIO_MODEL_PATH=FastVideo/MMAudio-large-44k-v2-Diffusers python examples/inference/basic/basic_mmaudio.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "limitations": ["The upstream checkpoint must be converted to Diffusers layout via scripts/checkpoint_conversion/convert_mmaudio_to_diffusers.py unless loaded from the FastVideo converted repo as done here."]
     },
     {
       "id": "matrix-game-2",
+      "family": "matrixgame",
+      "stage": "inference",
       "task": "Interactive world",
       "label": "Matrix Game 2.0",
+      "summary": "Generate an interactive-world sequence from the maintained Matrix Game 2.0 example.",
       "model": "FastVideo/Matrix-Game-2.0-Base-Distilled-Diffusers",
       "source": "examples/inference/basic/basic_matrixgame2.py",
-      "command": "python examples/inference/basic/basic_matrixgame2.py"
+      "command": "python examples/inference/basic/basic_matrixgame2.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "related": ["matrix-game-3-i2w"]
+    },
+    {
+      "id": "matrix-game-3-i2w",
+      "family": "matrixgame",
+      "stage": "inference",
+      "task": "Interactive world",
+      "label": "Matrix Game 3.0",
+      "summary": "Drive Matrix Game 3.0 from an input image plus prompt at 720p, three steps, from its maintained example.",
+      "model": "FastVideo/Matrix-Game-3.0-Base-Distilled-Diffusers",
+      "source": "examples/inference/basic/basic_matrixgame3.py",
+      "command": "python examples/inference/basic/basic_matrixgame3.py",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {"gpu_count": 1, "evidence": "source-configured"},
+      "evidence": "Source-backed",
+      "expected_artifact": "MP4 videos under video_samples_matrixgame3/",
+      "related": ["matrix-game-2"]
     }
   ]
 }

docs/assets/cookbook.js

@@ -1,57 +1,724 @@
 (() => {
-  let recipesPromise;
+  const PATTERN_CHARS = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789";
+  const PATTERN_LENGTH = 900;
+  const SCRAMBLE_MS = 140;
 
-  const loadRecipes = (url) => {
-    recipesPromise ||= fetch(url).then((response) => {
+  const loadRecipes = (url) =>
+    fetch(url).then((response) => {
       if (!response.ok) throw new Error(`HTTP ${response.status}`);
       return response.json();
     });
-    return recipesPromise;
+
+  const generatePattern = (length) => {
+    const chars = new Array(length);
+    for (let index = 0; index < length; index += 1) {
+      chars[index] = PATTERN_CHARS.charAt((Math.random() * PATTERN_CHARS.length) | 0);
+    }
+    return chars.join("");
+  };
+
+  const motionQuery = () =>
+    window.matchMedia("(hover: hover) and (pointer: fine) and (prefers-reduced-motion: no-preference)");
+
+  const initEvervault = (root) => {
+    root.querySelectorAll("[data-evervault]").forEach((visual) => {
+      if (visual.dataset.evervaultReady) return;
+      visual.dataset.evervaultReady = "true";
+      const noise = visual.querySelector("[data-cookbook-pattern]");
+      if (!noise) return;
+
+      const query = motionQuery();
+      if (!query.matches) return;
+
+      let rect = null;
+      let pointerX = 0;
+      let pointerY = 0;
+      let frame = 0;
+      let lastScramble = 0;
+      let patternReady = false;
+
+      const paint = (scramble) => {
+        visual.style.setProperty("--mouse-x", `${pointerX}px`);
+        visual.style.setProperty("--mouse-y", `${pointerY}px`);
+        if (scramble) noise.textContent = generatePattern(PATTERN_LENGTH);
+        frame = 0;
+      };
+
+      const onEnter = () => {
+        rect = visual.getBoundingClientRect();
+        if (!patternReady) {
+          noise.textContent = generatePattern(PATTERN_LENGTH);
+          patternReady = true;
+        }
+      };
+      const onMove = (event) => {
+        if (!rect) rect = visual.getBoundingClientRect();
+        pointerX = event.clientX - rect.left;
+        pointerY = event.clientY - rect.top;
+        if (frame) return;
+        frame = window.requestAnimationFrame(() => {
+          const now = performance.now();
+          const scramble = now - lastScramble >= SCRAMBLE_MS;
+          if (scramble) lastScramble = now;
+          paint(scramble);
+        });
+      };
+      const onLeave = () => {
+        if (frame) window.cancelAnimationFrame(frame);
+        frame = 0;
+        rect = null;
+        pointerX = 0;
+        pointerY = 0;
+        visual.style.setProperty("--mouse-x", "50%");
+        visual.style.setProperty("--mouse-y", "50%");
+      };
+      visual.addEventListener("mouseenter", onEnter);
+      visual.addEventListener("mousemove", onMove);
+      visual.addEventListener("mouseleave", onLeave);
+    });
+  };
+
+  let familyPopstate = null;
+  const bindFamilyPopstate = () => {
+    if (bindFamilyPopstate.bound) return;
+    bindFamilyPopstate.bound = true;
+    window.addEventListener("popstate", () => {
+      if (typeof familyPopstate === "function") familyPopstate();
+    });
+  };
+
+  const groupIdFor = (recipe) => recipe.group || recipe.id;
+
+  const gpuCountLabel = (hardware, runtimeId) => {
+    const count = hardware?.gpu_count;
+    if (count == null) return "";
+    if (runtimeId === "spark") return `${count} Spark${count === 1 ? "" : "s"}`;
+    return `${count} GPU${count === 1 ? "" : "s"}`;
+  };
+
+  const knobsFor = (recipe) => recipe.knobs || [];
+
+  const knobOptions = (knob) =>
+    knob.options.map((option) =>
+      option !== null && typeof option === "object" ? option : { value: option, label: String(option) },
+    );
+
+  const knobDefaultLabel = (knob) => {
+    const match = knobOptions(knob).find((option) => option.value === knob.default);
+    return match ? match.label : String(knob.default);
+  };
+
+  // Knob flags are always shown explicitly in the displayed command, even at
+  // their default value, so the command stays copy-pasteable and precise
+  // about what it runs -- not just "trust the script's own default".
+  const appendKnobFlags = (commandText, knobs, knobValues) => {
+    const flags = knobs
+      .filter((knob) => knobValues[knob.key] !== undefined)
+      .map((knob) => `${knob.flag} ${knobValues[knob.key]}`);
+    if (!flags.length) return commandText;
+    const lines = commandText.split("\n");
+    lines[lines.length - 1] = `${lines[lines.length - 1]} ${flags.join(" ")}`;
+    return lines.join("\n");
+  };
+
+  const runtimeFor = (recipe) => {
+    const platform = recipe.hardware?.platform || "cuda";
+    if (platform === "mlx") {
+      return {
+        id: "mlx",
+        label: "Apple Silicon · MLX",
+        hint:
+          recipe.hardware?.minimum_memory ||
+          [recipe.hardware?.accelerator, recipe.hardware?.system_memory].filter(Boolean).join(" · ") ||
+          "Memory not recorded",
+      };
+    }
+    if (platform === "mps") {
+      return {
+        id: "mps",
+        label: "Apple Silicon · MPS",
+        hint: recipe.hardware?.minimum_memory || recipe.hardware?.system_memory || "Memory not recorded",
+      };
+    }
+    const hardware = recipe.hardware || {};
+    if (hardware.device === "spark") {
+      return {
+        id: "spark",
+        label: "NVIDIA DGX Spark",
+        hint: "GB10 · 128 GB unified memory",
+      };
+    }
+    return {
+      id: "cuda",
+      label: "NVIDIA CUDA",
+      hint: hardware.accelerator
+        ? `${hardware.accelerator} · ${gpuCountLabel(hardware, "cuda")}`
+        : `${gpuCountLabel(hardware, "cuda")} configured · GPU model not recorded`,
+    };
+  };
+
+  const runtimeSummary = (recipe) => {
+    const runtime = runtimeFor(recipe);
+    const hardware = recipe.hardware || {};
+    if (runtime.id === "mlx") {
+      return [hardware.accelerator || "Apple Silicon", hardware.system_memory || hardware.minimum_memory, "MLX"]
+        .filter(Boolean)
+        .join(" · ");
+    }
+    if (runtime.id === "mps") {
+      return [hardware.accelerator || "Apple Silicon", hardware.system_memory || hardware.minimum_memory, "PyTorch MPS"]
+        .filter(Boolean)
+        .join(" · ");
+    }
+    if (runtime.id === "spark") {
+      return [hardware.accelerator || "NVIDIA GB10", gpuCountLabel(hardware, "spark")].filter(Boolean).join(" · ");
+    }
+    if (hardware.accelerator) return [hardware.accelerator, gpuCountLabel(hardware, runtime.id)].join(" · ");
+    return `NVIDIA CUDA · ${gpuCountLabel(hardware, "cuda")} configured · GPU model and VRAM not recorded`;
+  };
+
+  const renderHardwareEvidence = (container, badge, recipe) => {
+    const hardware = recipe.hardware || {};
+    const runtime = runtimeFor(recipe);
+    const isValidated = hardware.evidence === "validated";
+    container.classList.toggle("cookbook-hardware-state--verified", isValidated);
+    container.classList.toggle("cookbook-hardware-state--source", !isValidated);
+    badge.classList.toggle("cookbook-badge--verified", isValidated);
+    badge.classList.toggle("cookbook-badge--configured", !isValidated);
+    badge.textContent = isValidated ? "Recorded run" : "Source config";
+
+    const heading = document.createElement("strong");
+    heading.textContent = isValidated ? "Recorded hardware" : "Source configuration";
+    const details = document.createElement("span");
+
+    if (isValidated) {
+      const recorded = [
+        hardware.accelerator,
+        runtime.id === "mlx" || runtime.id === "mps" ? hardware.system_memory : gpuCountLabel(hardware, runtime.id),
+      ].filter(Boolean);
+      const statements = [`${recorded.join(" · ")}.`];
+      if (hardware.minimum_memory) statements.push(`Documented minimum: ${hardware.minimum_memory}.`);
+      if (hardware.peak_memory) statements.push(`Measured: ${hardware.peak_memory}.`);
+      if (!hardware.minimum_memory) statements.push("This recorded device is not a minimum requirement.");
+      details.textContent = ` ${statements.join(" ")}`;
+    } else if (runtime.id === "spark") {
+      details.textContent = ` NVIDIA DGX Spark · ${gpuCountLabel(hardware, "spark")}. GB10 has no FA4 / sm_100a VSA kernel; keep Triton VSA and FA4 off.`;
+    } else if (runtime.id === "cuda") {
+      details.textContent = ` NVIDIA CUDA · ${gpuCountLabel(hardware, "cuda")}. The source does not record the GPU model or VRAM.`;
+    } else {
+      details.textContent = ` ${runtime.label}. The source does not record a device or memory requirement.`;
+    }
+
+    container.replaceChildren(heading, details);
+    if (hardware.evidence_url) {
+      const evidenceLink = document.createElement("a");
+      evidenceLink.href = hardware.evidence_url;
+      evidenceLink.textContent = "View run evidence";
+      evidenceLink.setAttribute("aria-label", `View recorded hardware evidence for ${recipe.label}`);
+      container.append(" ", evidenceLink);
+    }
+  };
+
+  const compactLifecycle = (root) => {
+    const lifecycle = root.querySelector(".cookbook-lifecycle");
+    if (!lifecycle || lifecycle.dataset.compact) return;
+    lifecycle.dataset.compact = "true";
+    const stages = [...lifecycle.querySelectorAll(".cookbook-lifecycle__stage")];
+    const active = stages.find((stage) => stage.classList.contains("cookbook-lifecycle__stage--active"));
+    const planned = stages
+      .filter((stage) => stage !== active)
+      .map((stage) => stage.childNodes[0]?.textContent?.trim())
+      .filter(Boolean);
+    const summary = document.createElement("span");
+    summary.className = "cookbook-lifecycle__summary";
+    summary.textContent = `Next: ${planned.join(", ")}`;
+    lifecycle.replaceChildren(...(active ? [active] : []), summary);
+  };
+
+  const initFamilyBuilder = async (root) => {
+    const family = root.dataset.family;
+    if (!family) return;
+
+    compactLifecycle(root);
+
+    const modelOptions = root.querySelector("[data-cookbook-model-options]");
+    const hardwareOptions = root.querySelector("[data-cookbook-hardware-options]");
+    const description = root.querySelector("[data-cookbook-description]");
+    const label = root.querySelector("[data-cookbook-label]");
+    const model = root.querySelector("[data-cookbook-model]");
+    const task = root.querySelector("[data-cookbook-task]");
+    const hardwareValue = root.querySelector("[data-cookbook-gpus]");
+    const artifact = root.querySelector("[data-cookbook-artifact]");
+    const evidenceCell = root.querySelector("[data-cookbook-evidence]");
+    const source = root.querySelector("[data-cookbook-source]");
+    const modelLink = root.querySelector("[data-cookbook-model-link]");
+    const command = root.querySelector("[data-cookbook-command]");
+    const status = root.querySelector("[data-cookbook-status]");
+    const hardwareState = root.querySelector("[data-cookbook-hardware-state]");
+    const hardwareBadge = root.querySelector("[data-cookbook-hardware-badge]");
+    const count = root.querySelector("[data-cookbook-count]");
+    const result = root.querySelector(".cookbook-result");
+    const commandBlock = root.querySelector(".cookbook-command");
+    const servingPanel = root.querySelector("[data-cookbook-serving]");
+    const usage = root.querySelector("[data-cookbook-usage]");
+    const servingAvailability = root.querySelector("[data-cookbook-serving-availability]");
+    const knobsContainer = root.querySelector("[data-cookbook-knobs]");
+    const deviceRow = root.querySelector("[data-cookbook-device-row]");
+    const deviceOptions = root.querySelector("[data-cookbook-device-options]");
+    const deviceCaption = root.querySelector("[data-cookbook-device-caption]");
+
+    modelOptions.setAttribute("aria-label", "Recipe");
+    hardwareOptions.setAttribute("aria-label", "Runtime");
+
+    let recipes;
+    try {
+      ({ recipes } = await loadRecipes(root.dataset.recipes));
+    } catch (error) {
+      if (status) status.textContent = "Recipes could not be loaded. Use the maintained examples link below.";
+      console.error("Failed to load FastVideo cookbook recipes", error);
+      return;
+    }
+    if (!root.isConnected) return;
+
+    const familyRecipes = recipes.filter((recipe) => recipe.family === family);
+    if (!familyRecipes.length) return;
+
+    let servingProfiles = {};
+    let servingLoadFailed = false;
+    if (servingPanel && familyRecipes.some((recipe) => recipe.serving)) {
+      try {
+        const dataUrl = new URL(root.dataset.recipes, document.baseURI);
+        servingProfiles = await loadRecipes(new URL("cookbook-serving.json", dataUrl));
+      } catch (error) {
+        servingLoadFailed = true;
+        console.error("Failed to load FastVideo serving profiles", error);
+      }
+    }
+
+    const byId = new Map(familyRecipes.map((recipe) => [recipe.id, recipe]));
+    const groups = new Map();
+    familyRecipes.forEach((recipe) => {
+      const groupId = groupIdFor(recipe);
+      if (!groups.has(groupId)) groups.set(groupId, []);
+      groups.get(groupId).push(recipe);
+    });
+    if (count) count.textContent = `${familyRecipes.length} maintained recipes`;
+
+    modelOptions.replaceChildren();
+    groups.forEach((groupRecipes, groupId) => {
+      const representative = groupRecipes[0];
+      const option = document.createElement("button");
+      option.type = "button";
+      option.dataset.recipeGroup = groupId;
+      option.setAttribute("aria-pressed", "false");
+      const optionLabel = document.createElement("strong");
+      optionLabel.textContent = representative.group_label || representative.label;
+      const optionTask = document.createElement("span");
+      optionTask.textContent = representative.group_task || representative.task;
+      option.append(optionLabel, optionTask);
+      modelOptions.append(option);
+    });
+
+    const query = new URLSearchParams(window.location.search);
+    const requestedRecipe = query.get("recipe");
+    const defaultRecipeId = byId.has(root.dataset.defaultRecipe) ? root.dataset.defaultRecipe : familyRecipes[0].id;
+    let selectedRecipeId = requestedRecipe && byId.has(requestedRecipe) ? requestedRecipe : defaultRecipeId;
+    let selectedGroupId = groupIdFor(byId.get(selectedRecipeId));
+    let renderedRuntimeGroup = null;
+    // Keep previously shared local/openai links working after renaming workflows.
+    const workflow = (value) => ["local", "python"].includes(value) ? "python" : "server";
+    let usagePreference = workflow(query.get("use"));
+    let selectedClient = ["python", "javascript", "curl"].includes(query.get("client")) ? query.get("client") : "curl";
+    const clientDetails = servingPanel?.querySelector(".cookbook-serving__code");
+    if (clientDetails && query.has("client")) clientDetails.open = true;
+
+    const knobDefs = new Map();
+    familyRecipes.forEach((recipe) => knobsFor(recipe).forEach((knob) => {
+      if (!knobDefs.has(knob.key)) knobDefs.set(knob.key, knob);
+    }));
+    const knobValues = {};
+    knobDefs.forEach((knob, key) => {
+      const fromQuery = query.get(key);
+      const validValues = knobOptions(knob).map((option) => String(option.value));
+      const useQueryValue = fromQuery !== null && validValues.includes(fromQuery);
+      const raw = useQueryValue ? fromQuery : knob.default;
+      knobValues[key] = typeof knob.default === "number" ? Number(raw) : raw;
+    });
+
+    const renderKnobs = (recipe, hidden) => {
+      if (!knobsContainer) return;
+      const knobs = knobsFor(recipe);
+      const renderedKeys = [...knobsContainer.querySelectorAll("[data-knob-row]")].map((row) => row.dataset.knobRow);
+      if (renderedKeys.join(",") !== knobs.map((knob) => knob.key).join(",")) {
+        knobsContainer.replaceChildren();
+        knobs.forEach((knob) => {
+          const row = document.createElement("div");
+          row.className = "cookbook-selection-row";
+          row.dataset.knobRow = knob.key;
+          const labelWrap = document.createElement("div");
+          labelWrap.className = "cookbook-selection-row__label";
+          const strongLabel = document.createElement("strong");
+          strongLabel.textContent = knob.label;
+          const hintLabel = document.createElement("span");
+          hintLabel.textContent = knob.hint || "";
+          labelWrap.append(strongLabel, hintLabel);
+          const grid = document.createElement("div");
+          grid.className = "cookbook-option-grid cookbook-option-grid--hardware";
+          grid.setAttribute("role", "group");
+          grid.setAttribute("aria-label", knob.label);
+          knobOptions(knob).forEach((option) => {
+            const optionButton = document.createElement("button");
+            optionButton.type = "button";
+            optionButton.dataset.knobKey = knob.key;
+            optionButton.dataset.knobValue = String(option.value);
+            optionButton.setAttribute("aria-pressed", "false");
+            const optionLabel = document.createElement("strong");
+            optionLabel.textContent = option.label;
+            optionButton.append(optionLabel);
+            grid.append(optionButton);
+          });
+          row.append(labelWrap, grid);
+          knobsContainer.append(row);
+        });
+      }
+      knobsContainer.hidden = hidden || knobs.length === 0;
+      knobsContainer.querySelectorAll("button[data-knob-key]").forEach((optionButton) => {
+        const selected = String(knobValues[optionButton.dataset.knobKey]) === optionButton.dataset.knobValue;
+        optionButton.classList.toggle("cookbook-option--selected", selected);
+        optionButton.setAttribute("aria-pressed", String(selected));
+      });
+    };
+
+    const recipesForRuntime = (runtimeId, groupRecipes) =>
+      groupRecipes.filter((item) => runtimeFor(item).id === runtimeId);
+
+    const uniqueRuntimeIds = (groupRecipes) => {
+      const ids = [];
+      groupRecipes.forEach((item) => {
+        const id = runtimeFor(item).id;
+        if (!ids.includes(id)) ids.push(id);
+      });
+      return ids;
+    };
+
+    const pickRecipeForRuntime = (runtimeId, preferredCount, groupRecipes) => {
+      const siblings = recipesForRuntime(runtimeId, groupRecipes);
+      if (!siblings.length) return null;
+      if (preferredCount != null) {
+        const match = siblings.find((item) => item.hardware?.gpu_count === preferredCount);
+        if (match) return match;
+      }
+      return siblings.find((item) => item.hardware?.gpu_count === 1) || siblings[0];
+    };
+
+    const renderRuntimeOptions = () => {
+      const groupRecipes = groups.get(selectedGroupId) || [];
+      const runtimeIds = uniqueRuntimeIds(groupRecipes);
+      const renderedIds = [...hardwareOptions.querySelectorAll("[data-runtime-id]")].map((option) => option.dataset.runtimeId);
+      if (renderedIds.join(",") === runtimeIds.join(",")) return;
+      hardwareOptions.replaceChildren();
+      runtimeIds.forEach((runtimeId) => {
+        const representative = pickRecipeForRuntime(runtimeId, 1, groupRecipes);
+        const runtime = runtimeFor(representative);
+        const option = document.createElement("button");
+        option.type = "button";
+        option.dataset.runtimeId = runtime.id;
+        option.setAttribute("aria-pressed", "false");
+        const optionLabel = document.createElement("strong");
+        optionLabel.textContent = runtime.label;
+        const optionHint = document.createElement("span");
+        optionHint.textContent = runtime.hint;
+        option.append(optionLabel, optionHint);
+        hardwareOptions.append(option);
+      });
+    };
+
+    const renderDeviceOptions = (recipe) => {
+      if (!deviceRow || !deviceOptions) return;
+      const groupRecipes = groups.get(selectedGroupId) || [];
+      const runtime = runtimeFor(recipe);
+      const siblings = recipesForRuntime(runtime.id, groupRecipes)
+        .slice()
+        .sort((left, right) => (left.hardware?.gpu_count || 0) - (right.hardware?.gpu_count || 0));
+      const show = siblings.length > 1;
+      deviceRow.hidden = !show;
+      if (deviceCaption) {
+        deviceCaption.textContent = runtime.id === "spark" ? "1 Spark or a QSFP pair" : "GPU count for this runtime";
+      }
+      if (!show) {
+        deviceOptions.replaceChildren();
+        return;
+      }
+      const renderedIds = [...deviceOptions.querySelectorAll("[data-recipe-id]")].map((option) => option.dataset.recipeId);
+      if (renderedIds.join(",") !== siblings.map((item) => item.id).join(",")) {
+        deviceOptions.replaceChildren();
+        siblings.forEach((candidate) => {
+          const option = document.createElement("button");
+          option.type = "button";
+          option.dataset.recipeId = candidate.id;
+          option.setAttribute("aria-pressed", "false");
+          const optionLabel = document.createElement("strong");
+          optionLabel.textContent = gpuCountLabel(candidate.hardware, runtime.id);
+          const optionHint = document.createElement("span");
+          optionHint.textContent = runtime.id === "spark" && candidate.hardware?.gpu_count === 2
+            ? "Ray sequence parallel over QSFP"
+            : runtime.id === "spark"
+              ? "One GB10, local process"
+              : `${gpuCountLabel(candidate.hardware, runtime.id)} configured`;
+          option.append(optionLabel, optionHint);
+          deviceOptions.append(option);
+        });
+      }
+      deviceOptions.querySelectorAll("button").forEach((option) => {
+        const selected = option.dataset.recipeId === recipe.id;
+        option.classList.toggle("cookbook-option--selected", selected);
+        option.setAttribute("aria-pressed", String(selected));
+      });
+    };
+
+    let notes = root.querySelector("[data-cookbook-notes]");
+    if (!notes) {
+      notes = document.createElement("aside");
+      notes.className = "cookbook-recipe-notes";
+      notes.dataset.cookbookNotes = "";
+      notes.hidden = true;
+      result.insertBefore(notes, commandBlock);
+    }
+
+    const render = ({ groupChanged = false, historyMode = "replace" } = {}) => {
+      if (!root.isConnected) return;
+      if (!byId.has(selectedRecipeId)) selectedRecipeId = defaultRecipeId;
+      let recipe = byId.get(selectedRecipeId);
+      if (groupChanged || groupIdFor(recipe) !== selectedGroupId) {
+        const currentRuntime = runtimeFor(recipe).id;
+        const currentCount = recipe.hardware?.gpu_count;
+        const groupRecipes = groups.get(selectedGroupId) || [];
+        recipe = pickRecipeForRuntime(currentRuntime, currentCount, groupRecipes) || groupRecipes[0];
+        selectedRecipeId = recipe.id;
+      }
+
+      selectedGroupId = groupIdFor(recipe);
+      if (renderedRuntimeGroup !== selectedGroupId) {
+        renderRuntimeOptions();
+        renderedRuntimeGroup = selectedGroupId;
+      }
+      renderDeviceOptions(recipe);
+      const runtime = runtimeFor(recipe);
+      const profile = servingPanel && servingProfiles[recipe.id];
+      const useServer = Boolean(profile && usagePreference === "server");
+      // The measured local profile and the server config have separate evidence.
+      const activeRecipe = useServer ? { ...recipe, hardware: profile.hardware, evidence: "Source-backed" } : recipe;
+      const knobs = knobsFor(recipe);
+      renderKnobs(recipe, useServer);
+
+      if (usage) {
+        usage.querySelectorAll("[data-cookbook-mode]").forEach((option) => {
+          const selected = option.dataset.cookbookMode === (useServer ? "server" : "python");
+          option.disabled = option.dataset.cookbookMode === "server" && !profile;
+          option.classList.toggle("cookbook-option--selected", selected);
+          option.setAttribute("aria-pressed", String(selected));
+        });
+        servingAvailability.textContent = profile
+          ? "The playground and API clients share one server process. Both workflows can run on your own machine."
+          : servingLoadFailed
+            ? "Server examples could not be loaded. Open the H3 server guide below, or use Python directly."
+            : "This recipe uses Python directly. For the playground and API clients, choose FastH3 Preview with CUDA, MLX, or one Spark.";
+        servingPanel.hidden = !useServer;
+        commandBlock.hidden = useServer;
+        root.querySelector("[data-cookbook-python-note]").hidden = useServer;
+      }
+
+      if (useServer) {
+        const isMLX = profile.runtime === "mlx";
+        const isSpark = runtime.id === "spark";
+        servingPanel.querySelector("[data-cookbook-server-lifetime]").textContent = isMLX
+          ? "Start once, then change prompts in the playground or your app. MLX reuses its pipeline and prompt cache, but loads and releases model components between phases to limit unified-memory use. It does not keep all weights resident."
+          : isSpark
+            ? "Start once, then change prompts in the playground or your app. On a DGX Spark, lazy module load still reloads Qwen3-VL and the DiT between phases of each request, so later prompts are not a free hot cache."
+            : "Start once, then change prompts in the playground or your app. CUDA requests reuse the loaded model. The Python SDK can also reuse a generator within one process.";
+        servingPanel.querySelector("[data-cookbook-install-guide]").href = isMLX
+          ? "../../getting_started/installation/mps/#run-fasth3-preview"
+          : isSpark
+            ? "../../getting_started/installation/spark/"
+            : "../../getting_started/installation/gpu/";
+        servingPanel.querySelector("[data-cookbook-prepare]").hidden = !profile.prepare;
+        servingPanel.querySelector("[data-cookbook-server-prepare]").textContent = profile.prepare;
+        servingPanel.querySelector("[data-cookbook-server-install]").textContent = profile.install;
+        servingPanel.querySelector("[data-cookbook-server-command]").textContent = profile.command;
+        servingPanel.querySelector("[data-cookbook-health-command]").textContent = profile.health_command;
+        servingPanel.querySelector("[data-cookbook-playground]").href = profile.playground_url;
+        const client = profile.clients[selectedClient];
+        const filename = client.source.split("/").pop();
+        servingPanel.querySelector("[data-cookbook-client-install]").textContent = client.install;
+        const clientCode = servingPanel.querySelector("[data-cookbook-client-code]");
+        clientCode.className = `language-${selectedClient === "curl" ? "bash" : selectedClient}`;
+        clientCode.textContent = client.code;
+        servingPanel.querySelector("[data-cookbook-client-filename]").textContent = filename;
+        servingPanel.querySelector("[data-cookbook-client-source]").href = `https://github.com/hao-ai-lab/FastVideo/blob/main/${client.source}`;
+        const runner = { python: "python", javascript: "node", curl: "bash" }[selectedClient];
+        servingPanel.querySelector("[data-cookbook-client-run]").textContent = `Save as ${filename} and run ${runner} ${filename}. The MP4 is saved with the job ID as its filename.`;
+        servingPanel.querySelectorAll("[data-cookbook-client]").forEach((option) => {
+          option.setAttribute("aria-pressed", String(option.dataset.cookbookClient === selectedClient));
+        });
+      }
+
+      modelOptions.querySelectorAll("button").forEach((option) => {
+        const selected = option.dataset.recipeGroup === selectedGroupId;
+        option.classList.toggle("cookbook-option--selected", selected);
+        option.setAttribute("aria-pressed", String(selected));
+      });
+      hardwareOptions.querySelectorAll("button").forEach((option) => {
+        const selected = option.dataset.runtimeId === runtime.id;
+        option.classList.toggle("cookbook-option--selected", selected);
+        option.setAttribute("aria-pressed", String(selected));
+      });
+
+      description.textContent = useServer
+        ? `FastH3 Preview generates video with audio. This server profile uses the checked-in ${runtime.label} configuration.`
+        : recipe.summary;
+      label.textContent = useServer ? `${recipe.group_label || recipe.label} · Server` : recipe.label;
+      model.textContent = recipe.model;
+      task.textContent = recipe.task;
+      hardwareValue.textContent = runtimeSummary(activeRecipe);
+      if (artifact) artifact.textContent = useServer ? "MP4 with audio" : recipe.expected_artifact || "Not yet documented for this recipe.";
+      if (evidenceCell) {
+        evidenceCell.textContent = activeRecipe.evidence || "Source-backed";
+        evidenceCell.classList.toggle("cookbook-badge--verified", activeRecipe.evidence === "Verified");
+        evidenceCell.classList.toggle("cookbook-badge--source-backed", activeRecipe.evidence !== "Verified");
+      }
+      source.href = `https://github.com/hao-ai-lab/FastVideo/blob/main/${useServer ? profile.source : recipe.source}`;
+      source.textContent = useServer ? "View server configuration" : "Open example source";
+      modelLink.href = `https://huggingface.co/${recipe.model}`;
+      command.textContent = useServer ? recipe.command : appendKnobFlags(recipe.command, knobs, knobValues);
+
+      renderHardwareEvidence(hardwareState, hardwareBadge, activeRecipe);
+
+      const knobCaveats = useServer ? [] : knobs
+        .filter((knob) => String(knobValues[knob.key]) !== String(knob.default))
+        .map((knob) => `${knob.label} is set away from its recorded default (${knobDefaultLabel(knob)}). ` +
+          "The script accepts this value, but it has not been benchmarked here.");
+      const limitations = [...(useServer ? [
+        profile.runtime === "mlx"
+          ? "This MLX server config has no recorded hardware run. Measurements from the Python recipe are not server memory requirements. Only text-to-video/audio is wired; reference inputs and fast modes are not exposed here."
+          : runtime.id === "spark"
+            ? "This Spark server config has no recorded serving benchmark. Lazy module load reloads Qwen3-VL and the DiT between phases of each request. Compilation of the DiT is disabled."
+            : "This server config has no recorded serving benchmark. Compilation is disabled, unlike the measured Python performance profile.",
+        `${profile.sampling.width} × ${profile.sampling.height} · ${profile.sampling.num_frames} frames · ${profile.sampling.fps} fps. The server supplies these defaults; the client sends the model and prompt.`,
+        "Generation is serialized. Job metadata is held in memory and is lost when the server restarts.",
+      ] : recipe.limitations || []), ...knobCaveats];
+      notes.replaceChildren();
+      notes.hidden = limitations.length === 0;
+      if (limitations.length) {
+        const notesHeading = document.createElement("strong");
+        notesHeading.textContent = "Know before you run";
+        const notesList = document.createElement("ul");
+        limitations.forEach((item) => {
+          const listItem = document.createElement("li");
+          listItem.textContent = item;
+          notesList.append(listItem);
+        });
+        notes.append(notesHeading, notesList);
+      }
+
+      const nextQuery = new URLSearchParams(window.location.search);
+      nextQuery.set("recipe", recipe.id);
+      nextQuery.set("runtime", runtime.id);
+      const deviceSiblings = recipesForRuntime(runtime.id, groups.get(selectedGroupId) || []);
+      if (deviceSiblings.length > 1 && recipe.hardware?.gpu_count != null) {
+        nextQuery.set("gpus", String(recipe.hardware.gpu_count));
+      } else {
+        nextQuery.delete("gpus");
+      }
+      knobDefs.forEach((knob, key) => {
+        if (knobs.some((activeKnob) => activeKnob.key === key)) nextQuery.set(key, String(knobValues[key]));
+        else nextQuery.delete(key);
+      });
+      if (usage) {
+        nextQuery.set("use", useServer ? "server" : "python");
+        if (useServer && clientDetails?.open) nextQuery.set("client", selectedClient);
+        else nextQuery.delete("client");
+      }
+      const nextUrl = `${window.location.pathname}?${nextQuery.toString()}${window.location.hash}`;
+      if (historyMode === "push") window.history.pushState({}, "", nextUrl);
+      else if (historyMode === "replace") window.history.replaceState({}, "", nextUrl);
+      const modeSummary = useServer ? " with a persistent server" : "";
+      status.textContent = `${recipe.label} selected for ${runtime.label}${modeSummary}.`;
+    };
+
+    modelOptions.addEventListener("click", (event) => {
+      const option = event.target.closest("button[data-recipe-group]");
+      if (!option) return;
+      selectedGroupId = option.dataset.recipeGroup;
+      render({ groupChanged: true, historyMode: "push" });
+    });
+    hardwareOptions.addEventListener("click", (event) => {
+      const option = event.target.closest("button[data-runtime-id]");
+      if (!option) return;
+      const groupRecipes = groups.get(selectedGroupId) || [];
+      const currentCount = byId.get(selectedRecipeId)?.hardware?.gpu_count;
+      const nextRecipe = pickRecipeForRuntime(option.dataset.runtimeId, currentCount, groupRecipes);
+      if (!nextRecipe) return;
+      selectedRecipeId = nextRecipe.id;
+      selectedGroupId = groupIdFor(nextRecipe);
+      render({ historyMode: "push" });
+    });
+    deviceOptions?.addEventListener("click", (event) => {
+      const option = event.target.closest("button[data-recipe-id]");
+      if (!option) return;
+      selectedRecipeId = option.dataset.recipeId;
+      selectedGroupId = groupIdFor(byId.get(selectedRecipeId));
+      render({ historyMode: "push" });
+    });
+    usage?.addEventListener("click", (event) => {
+      const option = event.target.closest("button[data-cookbook-mode]");
+      if (!option || option.disabled) return;
+      usagePreference = option.dataset.cookbookMode;
+      render({ historyMode: "push" });
+    });
+    servingPanel?.addEventListener("click", (event) => {
+      const option = event.target.closest("button[data-cookbook-client]");
+      if (!option) return;
+      selectedClient = option.dataset.cookbookClient;
+      render({ historyMode: "push" });
+    });
+    knobsContainer?.addEventListener("click", (event) => {
+      const option = event.target.closest("button[data-knob-key]");
+      if (!option) return;
+      const knob = knobDefs.get(option.dataset.knobKey);
+      knobValues[option.dataset.knobKey] = typeof knob.default === "number"
+        ? Number(option.dataset.knobValue) : option.dataset.knobValue;
+      render({ historyMode: "push" });
+    });
+
+    render();
+
+    familyPopstate = () => {
+      if (!root.isConnected) return;
+      const nextQuery = new URLSearchParams(window.location.search);
+      const nextRecipe = nextQuery.get("recipe");
+      selectedRecipeId = nextRecipe && byId.has(nextRecipe) ? nextRecipe : defaultRecipeId;
+      selectedGroupId = groupIdFor(byId.get(selectedRecipeId));
+      usagePreference = workflow(nextQuery.get("use"));
+      selectedClient = ["python", "javascript", "curl"].includes(nextQuery.get("client")) ? nextQuery.get("client") : "curl";
+      if (clientDetails) clientDetails.open = nextQuery.has("client");
+      knobDefs.forEach((knob, key) => {
+        const fromQuery = nextQuery.get(key);
+        const validValues = knobOptions(knob).map((option) => String(option.value));
+        if (fromQuery !== null && validValues.includes(fromQuery)) {
+          knobValues[key] = typeof knob.default === "number" ? Number(fromQuery) : fromQuery;
+        }
+      });
+      render({ historyMode: "none" });
+    };
+    bindFamilyPopstate();
   };
 
   const init = () => {
-    document.querySelectorAll("[data-cookbook]").forEach(async (root) => {
+    initEvervault(document);
+    document.querySelectorAll("[data-cookbook][data-family]").forEach((root) => {
       if (root.dataset.initialized) return;
       root.dataset.initialized = "true";
-
-      const select = root.querySelector("[data-cookbook-recipe]");
-      const model = root.querySelector("[data-cookbook-model]");
-      const source = root.querySelector("[data-cookbook-source]");
-      const command = root.querySelector("[data-cookbook-command]");
-      const status = root.querySelector("[data-cookbook-status]");
-
-      try {
-        const { recipes } = await loadRecipes(root.dataset.recipes);
-        const byId = new Map(recipes.map((recipe) => [recipe.id, recipe]));
-        const groups = new Map();
-
-        select.replaceChildren();
-        recipes.forEach((recipe) => {
-          if (!groups.has(recipe.task)) {
-            const group = document.createElement("optgroup");
-            group.label = recipe.task;
-            groups.set(recipe.task, group);
-            select.append(group);
-          }
-          groups.get(recipe.task).append(new Option(recipe.label, recipe.id));
-        });
-
-        const render = () => {
-          const recipe = byId.get(select.value);
-          model.textContent = recipe.model;
-          source.textContent = recipe.source;
-          source.href = `https://github.com/hao-ai-lab/FastVideo/blob/main/${recipe.source}`;
-          command.textContent = recipe.command;
-          status.textContent = `${recipe.label} selected.`;
-        };
-
-        select.addEventListener("change", render);
-        select.disabled = false;
-        render();
-      } catch (error) {
-        status.textContent = "Recipes could not be loaded. Use the examples link below.";
-        console.error("Failed to load FastVideo cookbook recipes", error);
-      }
+      initFamilyBuilder(root);
     });
   };
 

docs/assets/logos/SOURCES.md

@@ -0,0 +1,26 @@
+# Cookbook logo sources
+
+All marks are official publisher assets from Hugging Face organization pages
+(vendored byte-for-byte from each org's public avatar). No imitation or
+redrawn logos are used. Families without a publisher-appropriate mark use a
+plain typographic tile instead — that tile is a UI placeholder, not a logo.
+
+| Asset | Source | Purpose |
+| --- | --- | --- |
+| `wan-ai.webp` | [Official Wan-AI Hugging Face organization avatar](https://huggingface.co/Wan-AI) | Wan family card and page header |
+| `ltx.webp` | [Official Lightricks Hugging Face organization avatar](https://huggingface.co/Lightricks) | LTX family card and page header |
+| `tencent-hunyuan.webp` | [Official Tencent Hunyuan Hugging Face organization avatar](https://huggingface.co/Tencent-Hunyuan) | Hunyuan and GameCraft cards, Hunyuan page header |
+| `nvidia.webp` | [Official NVIDIA Hugging Face organization avatar](https://huggingface.co/nvidia) | Cosmos and GEN3C cards, Cosmos page header |
+| `kandinsky.webp` | [Official Kandinsky Lab Hugging Face organization avatar](https://huggingface.co/kandinskylab) | Kandinsky 5 family card and page header |
+| `black-forest-labs.webp` | [Official Black Forest Labs Hugging Face organization avatar](https://huggingface.co/black-forest-labs) | FLUX family card and page header |
+| `minimax.webp` | [Official MiniMax Hugging Face organization avatar](https://huggingface.co/MiniMaxAI) | MiniMax H3 family card and page header |
+| `tongyi.webp` | [Official Tongyi MAI Hugging Face organization avatar](https://huggingface.co/Tongyi-MAI) | Z-Image family card and page header |
+| `zai.webp` | [Official Z.ai Hugging Face organization avatar](https://huggingface.co/zai-org) | GLM-Image family card and page header |
+| `stabilityai.webp` | [Official Stability AI Hugging Face organization avatar](https://huggingface.co/stabilityai) | Stable Diffusion and Stable Audio cards and page headers |
+| `meituan-longcat.webp` | [Official Meituan LongCat Hugging Face organization avatar](https://huggingface.co/meituan-longcat) | LongCat family card and page header |
+| `fastvideo.webp` | [Official FastVideo Hugging Face organization avatar](https://huggingface.co/FastVideo) | Matrix Game and MMAudio cards (converted weights published by this org), DreamX card |
+
+Typographic tiles (no vendored image): TurboDiffusion ("Turbo"), HY-World
+("HY"), LingBot ("LB"), MMAudio page header ("MMA"). These publishers have no
+single official mark appropriate for reuse in the catalog; add a licensed
+asset here if one becomes available.

docs/attention/vsa/index.md

@@ -6,6 +6,47 @@ Sparse attention mechanism selecting top-k blocks.
 
 VSA is included in the `fastvideo-kernel` package. See the [main Attention page](../index.md) for build instructions.
 
+## Apple Silicon (MiniMax H3 / FastH3)
+
+The native MLX runtime has an inference-only H3 VSA path that is separate from
+the CUDA `fastvideo-kernel` package:
+
+- **INT8 / INT6 / INT4** are **weight-only**. They cover linear matrices,
+  including `attn.to_gate_compress` when you convert with `--include-vsa`.
+  Attention Q/K/V stay BF16 (or the selected activation dtype). There is no
+  INT6 Q/K/V attention kernel.
+- **Dense-only checkpoints** (the default converter) drop the 50 gate
+  matrices and keep fused SDPA. They remain valid for dense inference.
+- **VSA-capable checkpoints** retain those gates, quantize them on the same
+  affine grid, and record `vsa.capable` in `mlx_h3_dit.json`. Runtime VSA is
+  still off until you pass `--vsa`.
+- **Tile sizes** 64 `(4, 4, 4)` and 256 `(4, 8, 8)`. Prefix keys can be
+  `exempt` or `compete`. `--vsa-dense-first-n-steps` and `--vsa-dense-layers`
+  force dense SDPA on the selected steps or blocks.
+- **`--vsa-impl auto`** uses the chunked gather+SDPA **reference** path.
+  `--vsa-impl simd` runs the SIMD-group Metal kernel (tile 64, head dim 128)
+  and falls back to reference on unsupported shapes or kernel failure. The
+  runtime executes a small kernel probe before use and remembers failures
+  for the process, so later blocks do not retry a broken backend.
+  It is not the default: 480p four-step generation is faster than reference
+  but does not yet match reference video. `--vsa-impl reference` is the same
+  as `auto`.
+
+See the [Apple Silicon guide](../../getting_started/installation/mps.md) for
+conversion and `mlx_fasth3.py` flags. Do not enable VSA on a dense-only
+checkpoint; reconvert with `--include-vsa` first.
+
+H3 uses fused MLX RMSNorm by default, including dense inference. This can
+change BF16 rounding relative to the older explicit normalization path.
+`FASTVIDEO_MLX_FAST_NORM` controls Wan normalization only.
+
+The generation report aggregates VSA statistics across all blocks and steps.
+`impl_counts` and `fallback_reasons` show mixed execution and fallback.
+`video_keep` is the mean number of selected video-key tiles per video query
+tile and head, including dense overrides. `achieved_sparsity` is the matching
+mean video-tile sparsity; in `compete` mode it measures actual selections,
+not the requested top-k budget. These are tile counts, not token-level FLOPs.
+
 ## Usage
 
 ```python

docs/contributing/ci_architecture.md

@@ -6,7 +6,8 @@ lives in [Testing](testing.md).
 
 ## Overview
 
-FastVideo splits validation across GitHub Actions, Buildkite, and Modal:
+FastVideo splits validation across GitHub Actions, Buildkite, Slinky Slurm,
+and Mergify:
 
 ```text
 PR opened or updated
@@ -16,17 +17,24 @@ PR opened or updated
   |     style, lint, type, spelling, Markdown, workflow syntax, filenames
   |
   |-- Tier 2: Fastcheck
-  |     Buildkite orchestrates Modal GPU jobs
-  |     path-filtered component and unit checks
+  |     Buildkite schedules six lanes on the Slinky Slurm cluster
+  |     encoder, VAE, transformer, kernel, unit, DreamVerse
   |
-  |-- /merge, /test full, or ready label
+  |-- /merge or ready label
         |
-        `-- Tier 3: Full Suite
-              Buildkite orchestrates Modal GPU jobs
-              path-filtered integration, SSIM, training, eval, and performance checks
+        `-- Tier 3: Change-aware merge gate
+              trusted base-branch planner classifies the complete PR diff
+              Buildkite adds only relevant integration, quality, training,
+              API, or performance lanes on Slinky Slurm
               |
               pass -> Mergify squash-merges when all merge conditions pass
               fail -> fix, push, and re-run
+
+  |-- /test full
+        `-- Explicit all-20-lane diagnostic run
+
+  `-- weekly schedule on main
+        `-- Complete four-GPU SSIM matrix
 ```
 
 CI is not one monolithic job:
@@ -34,10 +42,31 @@ CI is not one monolithic job:
 - GitHub Actions owns pre-commit, slash-command handling, aggregate status
   updates, docs deployment, image builds, package publishing, and community
   automations.
-- Buildkite owns the GPU test pipeline and path filtering.
-- Modal owns the actual GPU execution environment for test jobs.
+- Buildkite owns the GPU test graph, statuses, and trusted dispatch control
+  plane. Its agent runs on the Slurm login plane; it does not execute test
+  payloads.
+- Slinky Slurm is the only active CI compute backend. A host-owned dispatcher
+  leases GPUs from a persistent four-GPU allocation and runs each lane in an
+  isolated Enroot container at the immutable PR SHA.
 - Mergify owns merge protection, labeling, and the final squash merge.
 
+The old files under `fastvideo/tests/modal/` are retained as dormant manual
+rollback code. `.buildkite/scripts/pr_test.sh` rejects Buildkite invocations,
+and no pipeline or slash-command route calls Modal.
+
+Three Buildkite entry pipelines share the validated graph:
+
+| Pipeline | Trigger | Scope |
+|---|---|---|
+| `pr-fastcheck` | Automatic pull-request webhook | Six Fastcheck lanes |
+| `ci` | `/merge`, `ready`, schedules, and `/test` API builds | Change-aware merge gates, scheduled SSIM, explicit Full Suite, Fastcheck reruns, or one direct lane |
+| `fastvideo-performance-lane` | Weekly scheduler | Direct performance lane |
+
+Each entry pipeline starts with the same trusted `pipeline-upload` job on the
+`ci-runner` queue. The `ci` pipeline's incoming GitHub webhook is disabled;
+otherwise it would duplicate the automatic `pr-fastcheck` build. API and
+scheduled builds continue to work with webhook processing disabled.
+
 ## CI Tiers
 
 ### Tier 1: Pre-commit
@@ -70,34 +99,37 @@ debugging a hook implementation.
 | Attribute | Value |
 |---|---|
 | Triggered by | Buildkite PR builds with `TEST_SCOPE=fastcheck` or unset |
-| Runner | Buildkite agent that launches Modal GPU jobs |
+| Compute | Slinky Slurm (`ci-runner` queue) |
 | Definition | `.buildkite/pipeline.yml` |
-| Entrypoint | `.buildkite/scripts/pr_test.sh` -> `fastvideo/tests/modal/pr_test.py` |
+| Entrypoint | Trusted host driver -> `.buildkite/scripts/unit_test.sh` or `.buildkite/scripts/lanes/*.sh` |
 
-Fastcheck uses Buildkite's `monorepo-diff` plugin. Jobs whose watched paths did
-not change are skipped and do not block the aggregate `fastcheck-passed`
-status.
+Fastcheck always schedules these six lanes: encoder, VAE, transformer, custom
+kernels, unit tests, and DreamVerse. Static steps replace the former
+host-side path-filter plugin: the login plane never checks out or executes PR
+code.
 
-| Buildkite label | `TEST_TYPE` | Main watched paths |
-|---|---|---|
-| Encoder Tests | `encoder` | `fastvideo/models/encoders/**`, `fastvideo/models/loader/**`, `fastvideo/tests/encoders/**`, `pyproject.toml`, `docker/Dockerfile` |
-| VAE Tests | `vae` | `fastvideo/models/vaes/**`, `fastvideo/models/loader/**`, `fastvideo/tests/vaes/**`, `pyproject.toml`, `docker/Dockerfile` |
-| Transformer Tests | `transformer` | `fastvideo/models/dits/**`, `fastvideo/models/loader/**`, `fastvideo/tests/transformers/**`, `fastvideo/layers/**`, `fastvideo/attention/**`, `pyproject.toml`, `docker/Dockerfile` |
-| Kernel Tests | `kernel_tests` | `fastvideo-kernel/**`, `pyproject.toml`, `docker/Dockerfile` |
-| Unit Tests | `unit_test` | `fastvideo/**`, `.buildkite/**`, `.github/**`, `pyproject.toml`, `docker/Dockerfile` |
-| DreamVerse App Tests | `dreamverse_app` | `apps/dreamverse/**`, `pyproject.toml` |
-
-### Tier 3: Full Suite
+### Tier 3: Change-Aware Merge Gate
 
 | Attribute | Value |
 |---|---|
-| Triggered by | `/merge`, adding `ready`, `/test full`, or a new push to a PR that already has `ready` |
-| Runner | Buildkite agent that launches Modal GPU jobs |
+| Triggered by | `/merge`, adding `ready`, or a new push to a PR that already has `ready` |
+| Compute | Slinky Slurm only (`ci-runner` queue) |
 | Definition | `.buildkite/pipeline.yml` |
-| Entrypoint | `.buildkite/scripts/pr_test.sh` -> `fastvideo/tests/modal/pr_test.py` |
+| Entrypoint | `/opt/fastvideo-ci-runner/run-ci` (`run-unit` is a compatibility wrapper) |
 
-Full Suite is also path-filtered. It validates broader behavior before Mergify
-can merge a PR.
+Fastcheck is the universal six-lane baseline. The merge gate does not repeat
+those jobs: it classifies every changed path and adds only the relevant lanes
+from the fourteen-lane integration set below. Selected jobs are hard gates;
+there are no soft-fail hardware lanes. A documentation-only PR can therefore
+finish its merge build after the trusted uploader, while model-family changes
+typically add focused golden-gate and SSIM files and a training-only change
+adds only its owning training lane.
+
+`.github/scripts/plan_merge_ci.py` is the canonical path policy. It runs from
+the immutable base SHA under `pull_request_target`; PR code is never executed
+on the GitHub runner. The changed-file list includes both sides of renames. An
+API failure, truncated response, empty list, unknown build input, or unknown
+source path fails closed to all fourteen integration lanes.
 
 A `ready`-labeled PR does not hit Buildkite immediately:
 `ci-trigger-full-suite.yml` first runs `.github/scripts/gate_full_suite.sh`,
@@ -106,25 +138,118 @@ head. A red cheap check blocks the suite (fail closed; the next push re-arms
 it), while a GitHub outage or a >25 min wait lets it run anyway (fail open).
 `/test full` bypasses the gate.
 
-| Buildkite label | `TEST_TYPE` | Main watched paths |
-|---|---|---|
-| SSIM Tests | `ssim` | `fastvideo/**/*.py`, `pyproject.toml`, `docker/Dockerfile` |
-| LoRA Inference Tests | `inference_lora` | LoRA tests, loader, transformer tests, pipelines, LoRA layers |
-| LoRA Extraction Tests | `lora_extraction` | LoRA extraction scripts/tests, loader, training utilities, LoRA layers |
-| Training Tests | `training` | `fastvideo/**`, `pyproject.toml`, `docker/Dockerfile` |
-| Distillation DMD Tests | `distillation_dmd` | `fastvideo/training/*distillation_pipeline.py` |
-| Self-Forcing Tests | `self_forcing` | self-forcing distillation pipeline and tests |
-| LoRA Training Tests | `training_lora` | `fastvideo/**`, `pyproject.toml`, `docker/Dockerfile` |
-| Training Tests VSA | `training_vsa` | `fastvideo/**`, `fastvideo-kernel/**`, `pyproject.toml`, `docker/Dockerfile` |
-| Inference Tests VMoBA | `inference_vmoba` | `fastvideo-kernel/**`, `fastvideo/attention/backends/vmoba.py` |
-| Performance Tests | `performance` | DiTs, pipelines, attention, layers, worker, entrypoints, performance tests/configs |
-| API Server Tests | `api_server` | OpenAI entrypoints, serve CLI, OpenAI API integration test |
-| Train Framework Tests | `train_framework` | `fastvideo/train/**`, train model/method tests, model loader, DiTs |
-| Eval Metrics Tests | `eval` | `fastvideo/eval/**`, `fastvideo/tests/eval/**`, `pyproject.toml`, `docker/Dockerfile` |
+The complete static graph remains available through `/test full`; path
+selection never deletes or dynamically invents a Buildkite step.
+
+Integration steps depend on `golden-gate`. A failed selected golden prevents
+the expensive downstream jobs from starting; `/test full` still selects all
+twenty lanes. A condition-skipped golden satisfies the dependency, so direct
+lane reruns, scheduled SSIM, and training-only merge plans keep their existing
+meaning. This follows Buildkite's
+[conditional dependency rules](https://buildkite.com/docs/pipelines/configure/depends-on).
+No dependency allows failures. The trusted uploader accepts either the old
+complete graph or the complete golden-first graph during rollout, and rejects
+partial or arbitrary dependency changes.
+
+The six automatic Fastcheck lanes are unchanged. Within VAE and transformer
+lanes, small Wan goldens run before independent component parity. Wan paths
+select matching VAE, dense/trajectory, or causal-cache goldens; shared Wan
+config and pipeline wiring select all four. Shared runtime changes continue
+to select broader coverage. The existing block references retain their exact
+environment identity; new tensor gates distinguish the effective FA2/FA4
+switch and VAE gates do not depend on an unused attention backend.
+
+| Lane | Public `TEST_TYPE` | GPUs | Typical merge trigger |
+|---|---|---:|---|
+| Encoder | `encoder` | 1 | Universal Fastcheck |
+| VAE | `vae` | 1 | Universal Fastcheck |
+| Transformer | `transformer` | 1 | Universal Fastcheck |
+| Kernel | `kernel_tests` | 1 | Universal Fastcheck |
+| Unit | `unit_test` | 1 | Universal Fastcheck |
+| DreamVerse | `dreamverse_app` | 1 | Universal Fastcheck |
+| Golden gate | `golden_gate` | 1 | Model, pipeline, attention, layer, or output changes |
+| SSIM | `ssim` | 4 | Matching model/SSIM paths; focused files when possible |
+| LoRA inference | `inference_lora` | 1 | LoRA inference/shared LoRA paths |
+| LoRA extraction | `lora_extraction` | 1 | LoRA extraction/shared LoRA paths |
+| Vanilla training | `training` | 4 | Legacy vanilla/shared training paths |
+| DMD distillation | `distillation_dmd` | 2 | DMD/shared training paths |
+| Self-forcing | `self_forcing` | 2 | Self-forcing/shared training paths |
+| LoRA training | `training_lora` | 2 | LoRA/shared training paths |
+| VSA training | `training_vsa` | 2 | VSA/shared training paths |
+| VMoBA inference | `inference_vmoba` | 1 | VMoBA backend/config paths |
+| Performance | `performance` | 2 | Performance tests or benchmark policy |
+| API server | `api_server` | 1 | API, worker, or server entrypoint paths |
+| Modular train framework | `train_framework` | 1 | `fastvideo/train/` and its tests |
+| Eval metrics | `eval` | 1 | `fastvideo/eval/` and its tests |
+
+Golden-gate and SSIM selections are basenames, not arbitrary pytest arguments.
+The private host checks the comma-separated allowlist before staging, and the
+container checks it again before invoking pytest. Shared quality-harness
+changes still run the complete owning lane. The full SSIM matrix also runs on
+`main` every Sunday at 05:00 UTC through `ci-scheduled-ssim.yml`; `/test ssim`
+and `/test full` remain available for deliberate complete runs.
+
+The four Buildkite workers may accept multiple jobs concurrently. The
+agent-owned lease broker packs their requested GPU counts onto one persistent
+four-GPU Slurm allocation and waits when capacity is full. A four-GPU lane
+such as SSIM or vanilla training owns the whole tray; two two-GPU lanes or up
+to four one-GPU lanes can overlap without sharing devices. SSIM and vanilla
+training also share the `fastvideo/slinky/whole-tray` Buildkite concurrency
+group. That keeps the second whole-tray lane in Buildkite instead of consuming
+an agent and its command timeout while the first lane waits for four free GPUs.
+Because packed Enroot containers share the node network namespace, each GPU
+lease receives its own 100-port rendezvous range. Tests preserve the
+runner-assigned `MASTER_PORT`, and parallel SSIM tasks use distinct offsets
+inside that range.
 
 See [Performance Benchmarks](performance_benchmarks.md) for the performance
 lane's thresholds, rolling baseline, artifacts, and reseeding process.
 
+## `/merge` Request Flow
+
+The Buildkite agent and Slurm have deliberately separate responsibilities:
+
+```text
+/merge PR comment
+  -> GitHub verifies write permission and refreshes the `ready` label
+  -> base-branch `ci-trigger-full-suite` workflow fetches the PR file list,
+     computes MERGE_TEST_PLAN plus focused golden/SSIM basenames, and gates on
+     cheap checks
+  -> sends the PR SHA to pipeline `ci` with TEST_SCOPE=merge and FULL_SUITE=true
+  -> trusted `pipeline-upload` job on queue `ci-runner`
+       fetch exact-SHA .buildkite/pipeline.yml
+       normalize + validate the complete static 20-lane policy and conditions
+       upload static Buildkite steps
+  -> each accepted step reaches the host policy hook
+       validate org/repo/SHA/ref/step key/command/scope/timeout
+       skip checkout on the login plane
+       stage a mode-0600 request on Lustre
+       lease 1-4 GPUs and attach an `srun` step to the Slinky tray
+  -> Enroot worker container
+       clone and verify the exact PR SHA
+       install the lane's project extras and cached kernel
+       run the repository-owned lane script
+       write numeric exit status and approved artifacts
+  -> trusted host returns that status to Buildkite
+  -> Buildkite publishes `full-suite-passed` only when every selected lane passes
+```
+
+The pipeline uploader and dispatcher run on the Slurm login plane, but those
+are control-plane operations only. Python tests, model loading, CUDA kernels,
+Node/Playwright checks, inference, training, SSIM generation, and performance
+benchmarks all execute in Slurm allocations.
+
+PR-controlled values never become host commands. The host policy accepts only
+the pinned pipeline uploader or a known lane tuple. It rejects plugins,
+artifact globs, shell injection variables, non-immutable commits, and unknown
+commands before checkout. Hugging Face credentials are added only for lanes
+that declare them, passed through a mode-0600 request file, and removed before
+the PR payload starts. Active training lanes keep W&B offline and do not stage
+a W&B credential. The ARM64 image includes the pinned FA4 CuTe overlay validated
+on GB200. SSIM opts into FA4 to preserve its reference-video numerics; lanes
+with FA2 baselines keep `FASTVIDEO_FA4=0`. Performance artifacts are relayed
+afterward by the trusted host from an allowlisted directory and extension set.
+
 ## Slash Commands
 
 Slash commands are handled by `.github/workflows/ci-slash-commands.yml`.
@@ -132,7 +257,7 @@ Repository write permission is required.
 
 | Command | Effect |
 |---|---|
-| `/merge` | Adds `ready` and triggers Full Suite for the PR head branch. |
+| `/merge` | Adds `ready` and triggers the path-aware merge gate for the PR head. |
 | `/test full` | Runs the whole Full Suite with `TEST_SCOPE=full`. |
 | `/test fastcheck` | Runs the whole Fastcheck suite with `TEST_SCOPE=fastcheck`. |
 | `/test pre-commit` | Re-runs the pre-commit workflow on the PR merge ref. |
@@ -149,6 +274,7 @@ Valid direct test names:
 | `/test unit` | `unit_test` |
 | `/test dreamverse` | `dreamverse_app` |
 | `/test ssim` | `ssim` |
+| `/test golden-gate` | `golden_gate` |
 | `/test training` | `training` |
 | `/test lora-inference` | `inference_lora` |
 | `/test lora-training` | `training_lora` |
@@ -162,12 +288,18 @@ Valid direct test names:
 | `/test train-framework` | `train_framework` |
 | `/test eval` | `eval` |
 
+The temporary `<name>-ci` spellings remain accepted as compatibility aliases;
+they select the same Slurm lane and do not identify a second backend.
+
 When a direct test completes successfully, Buildkite posts
 `direct-test-completed`. `.github/workflows/ci-aggregate-status.yml` then reads
 the latest Buildkite statuses for the commit and updates `fastcheck-passed` or
 `full-suite-passed` if all jobs in that group are green.
 
-Skipped path-filtered jobs have no status entry and do not block the aggregate.
+Buildkite label emojis define the status namespace used by that aggregation:
+`:microscope:` is reserved for the six Fastcheck lanes, while Full-Suite-only
+lanes use `:test_tube:` or `:bar_chart:`. Each active lane has exactly one
+label and therefore one status context.
 
 ## Merge Protection
 
@@ -177,7 +309,7 @@ Mergify enforces these conditions before it squash-merges to `main`:
 |---|---|
 | `check-success~=pre-commit` | Tier 1 passed. |
 | `check-success=fastcheck-passed` | All triggered Fastcheck jobs passed. |
-| `check-success=full-suite-passed` | All triggered Full Suite jobs passed. |
+| `check-success=full-suite-passed` | The selected merge gate or explicit Full Suite passed. |
 | `#approved-reviews-by>=1` | At least one approving review. |
 | Valid title regex | PR title starts with an accepted `[type]` tag. |
 | `label=ready` | The PR has entered the merge flow. |
@@ -226,29 +358,40 @@ Process labels:
 
 | Label | Who sets it | Meaning |
 |---|---|---|
-| `ready` | `/merge` or maintainer action | Triggers/keeps Full Suite active and enables auto-merge. |
+| `ready` | `/merge` or maintainer action | Triggers/keeps the change-aware merge gate active and enables auto-merge. |
 | `needs-rebase` | Mergify | PR has merge conflicts. |
 | `do-not-merge` | Maintainer | Blocks merge regardless of CI status. |
 
-## Modal Test Entrypoints
+## Slurm Lane Entrypoints
 
-All Buildkite test jobs go through `.buildkite/scripts/pr_test.sh`, which:
+Every active test selection lives in `.buildkite/scripts/unit_test.sh` or a
+focused `.buildkite/scripts/lanes/<lane>.sh`. The private, agent-owned lane
+table binds each internal `*_ci` type to that script, its GPU count, wall-clock
+limit, dependency extras, kernel-build policy, secrets, and artifacts. The
+internal suffix is an implementation detail; there is only one active backend.
 
-1. Reads Buildkite secrets for Modal, Hugging Face, and W&B when needed.
-2. Selects a Modal function based on `TEST_TYPE`.
-3. Passes Buildkite metadata into the Modal container.
-4. Runs the selected test command from `fastvideo/tests/modal/pr_test.py` or
-   `fastvideo/tests/modal/ssim_test.py`.
-5. Uploads performance artifacts for `TEST_TYPE=performance`.
+SSIM uses `fastvideo/tests/ssim/ci_runner.py` inside a single four-GPU lease.
+It discovers `REQUIRED_GPUS` and `*_MODEL_TO_PARAMS` with AST parsing, then
+packs independent pytest subprocesses across the visible GPUs with fail-fast
+termination. Performance writes reports to a host-mounted artifact directory;
+the host uploads only `.md`, `.html`, `.json`, and `.csv` files after the
+container exits.
+
+The Modal launchers remain in the repository for manual rollback archaeology,
+but they are not CI entrypoints. `pr_test.sh` rejects Buildkite calls and needs
+`FASTVIDEO_ENABLE_LEGACY_MODAL_CI=1` even for a local manual invocation.
 
 If you add a new CI test category:
 
-1. Add the Modal function in `fastvideo/tests/modal/pr_test.py` or a focused
-   companion module.
-2. Add the `TEST_TYPE` case in `.buildkite/scripts/pr_test.sh`.
-3. Add the Buildkite direct-test step and any Fastcheck/Full Suite path filters
-   in `.buildkite/pipeline.yml`.
-4. Add or update the `/test` mapping in `.github/workflows/ci-slash-commands.yml`.
+1. Add an executable `.buildkite/scripts/lanes/<lane>.sh` containing the test
+   payload only.
+2. Add the static Buildkite step in `.buildkite/pipeline.yml`, its source/test
+   ownership in `.github/scripts/plan_merge_ci.py`, and the `/test` mapping in
+   `.github/workflows/ci-slash-commands.yml`.
+3. Extend `fastvideo/tests/contract/test_ci_test_collection.py`,
+   `test_merge_ci_plan.py`, and the trusted private runner's lane table plus
+   uploader policy in the same rollout.
+4. Validate on the target GB200 hardware before making the lane a merge gate.
 5. Document the lane here and link any domain-specific authoring guide.
 
 ## CD And Release Workflows
@@ -282,13 +425,18 @@ the explicit `py3.12-cuda13.0.0-latest` tag. This publication policy does not
 change the unparameterized `docker/Dockerfile` build defaults, which remain CUDA
 13 and `cu130`.
 
-Published amd64 development images keep their configured Hopper kernel wheel
-installed and also carry an immutable SM89 wheel under
-`/opt/fastvideo-kernel-prebuilt`. Modal PR and SSIM jobs select the exact
-source, ABI, and GPU-architecture match from that directory, so L40S jobs reuse
-the trusted image artifact while kernel-changing PRs still build locally. Once
-a kernel or artifact-key change reaches `main`, the image workflow republishes
-the matching trusted artifact before later jobs consume the updated image tag.
+Published development images carry architecture-specific kernel wheels under
+`/opt/fastvideo-kernel-prebuilt`. The Slurm worker selects the exact source,
+ABI, and GPU-architecture match, so normal lanes reuse the trusted artifact
+while kernel-changing PRs still build locally. Once a kernel or artifact-key
+change reaches `main`, the image workflow republishes the matching artifact
+before later jobs consume the updated image pin.
+
+The same workflow publishes a single-architecture ARM64, CUDA 13, SM100 image
+for the self-hosted CI runner under the
+`py3.12-cuda13.0.0-sm100-{latest,sha-*}` tags. It carries the matching prebuilt
+kernel wheel so runner jobs can validate and install the exact source and ABI
+match instead of recompiling it in every lane.
 
 The optional Dreamverse matrix builds backend and UI images for CUDA 12.6 and
 CUDA 13 on `amd64`. Dreamverse remains `amd64`-only because its FA4 dependency
@@ -320,15 +468,17 @@ The reusable implementation lives in
 | `.github/mergify.yml` | Merge protection, PR title validation, PR labels, conflict labels, auto-merge |
 | `.github/workflows/ci-precommit.yml` | Tier 1 pre-commit |
 | `.github/workflows/ci-slash-commands.yml` | `/merge` and `/test` handling |
-| `.github/workflows/ci-trigger-full-suite.yml` | Full Suite trigger for `ready` PRs and new pushes to ready PRs |
-| `.github/workflows/ci-aggregate-status.yml` | Aggregate Fastcheck/Full Suite commit statuses |
-| `.buildkite/pipeline.yml` | Buildkite test graph and path filters |
-| `.buildkite/scripts/pr_test.sh` | Buildkite-to-Modal test dispatcher |
-| `fastvideo/tests/modal/pr_test.py` | Modal functions for most GPU CI lanes |
-| `fastvideo/tests/modal/ssim_test.py` | Modal functions and partitioning for SSIM |
+| `.github/workflows/ci-trigger-full-suite.yml` | Change-aware merge-gate trigger for `ready` PRs and new pushes |
+| `.github/workflows/ci-scheduled-ssim.yml` | Weekly complete SSIM trigger on `main` |
+| `.github/scripts/plan_merge_ci.py` | Trusted changed-path to integration-lane planner |
+| `.github/workflows/ci-aggregate-status.yml` | Aggregate Fastcheck and explicit Full Suite direct-rerun statuses |
+| `.buildkite/pipeline.yml` | Static 20-lane Slurm Buildkite graph |
+| `.buildkite/scripts/unit_test.sh`, `.buildkite/scripts/lanes/*.sh` | Active Slurm lane payloads |
+| `fastvideo/tests/ssim/ci_runner.py` | Four-GPU Slurm SSIM scheduler |
+| `.buildkite/scripts/pr_test.sh`, `fastvideo/tests/modal/*.py` | Dormant manual Modal rollback path (disabled in Buildkite) |
 | `.buildkite/performance-benchmarks/tests/*.json` | Performance benchmark configs and thresholds |
 | `.github/workflows/infra-docs.yml` | Docs build and GitHub Pages deploy |
-| `.github/workflows/infra-build-image.yml` | Automatic CUDA matrix and manual Docker image builds |
+| `.github/workflows/infra-build-image.yml` | CUDA matrix, CI runner image, and manual Docker image builds |
 | `.github/workflows/publish-fastvideo.yml` | FastVideo PyPI publishing |
 | `.github/workflows/publish-kernel.yml` | FastVideo kernel PyPI publishing |
 | `.github/workflows/publish-comfyui.yml` | ComfyUI registry publishing |

docs/contributing/coding_agents.md

@@ -43,6 +43,8 @@ FastVideo maps a Diffusers-style repo into a pipeline like:
 - `fastvideo/models/*`: model implementations (DiT, VAE, encoders, upsamplers).
 - `fastvideo/configs/models/*`: arch configs and `param_names_mapping` for
   weight name translation.
+- `fastvideo/models/wan/`: Wan's dense transformer, VAE, and component configs
+  live together. The old Wan modules remain compatibility re-exports.
 - `fastvideo/configs/pipelines/*`: pipeline wiring (component classes + names).
 - `fastvideo/api/sampling_param.py`: runtime sampling parameters.
 - `fastvideo/pipelines/basic/*`: end-to-end pipeline logic built from stages.
@@ -209,7 +211,7 @@ class OfficialWanTransformer(torch.nn.Module):
     def forward(self, x):
         return self.patch_embedding(x)
 
-# FastVideo model (simplified) in fastvideo/models/dits/wanvideo.py
+# FastVideo model (simplified) in fastvideo/models/wan/transformer.py
 class PatchEmbed(torch.nn.Module):
     def __init__(self):
         super().__init__()
@@ -227,7 +229,7 @@ class WanTransformer3DModel(torch.nn.Module):
         return self.patch_embedding(x)
 
 # Mapping defined in a config (simplified; see the real mapping in
-# fastvideo/configs/models/dits/wanvideo.py)
+# fastvideo/models/wan/config.py)
 param_names_mapping = {
     r"^patch_embedding\.(.*)$": r"patch_embedding.proj.\1",
     r"^blocks\.(\d+)\.attn1\.to_q\.(.*)$": r"blocks.\1.to_q.\2",
@@ -275,7 +277,7 @@ Mapping steps:
   - Instantiate the FastVideo DiT (`WanTransformer3DModel`) and compare
     its `state_dict().keys()` to the official keys.
   - Update `param_names_mapping` in
-    fastvideo/configs/models/dits/wanvideo.py to resolve missing/unexpected keys.
+    fastvideo/models/wan/config.py to resolve missing/unexpected keys.
   - Use `load_state_dict(strict=False)` during iteration to surface mismatches.
 ```
 
@@ -466,19 +468,24 @@ The Wan2.1 T2V 1.3B Diffusers pipeline is a good “standard” example for
 FastVideo integration.
 
 1. Verify model config + mapping.
-   - DiT mapping: `fastvideo/configs/models/dits/wanvideo.py`
-   - VAE: `fastvideo/models/vaes/wanvae.py`
+   - DiT: `fastvideo/models/wan/transformer.py`
+   - DiT mapping: `fastvideo/models/wan/config.py`
+   - VAE: `fastvideo/models/wan/vae.py`
+   - VAE config: `fastvideo/models/wan/vae_config.py`
    - Text encoder: `fastvideo/models/encoders/t5.py`
 
 2. Parity test the core components.
+   - Start with `bash scripts/validate_wan.sh all`: contracts and tiny goldens.
    - Example tests: `fastvideo/tests/transformers/test_wanvideo.py`,
      `fastvideo/tests/vaes/test_wan_vae.py`,
      `fastvideo/tests/encoders/test_t5_encoder.py`
 
 3. Pipeline wiring.
    - Pipeline: `fastvideo/pipelines/basic/wan/wan_pipeline.py`
-   - Pipeline config: `fastvideo/configs/pipelines/wan.py`
-   - Sampling defaults: `fastvideo/pipelines/basic/wan/profiles.py`
+   - Denoising and first-frame preparation: `fastvideo/pipelines/basic/wan/stages/`
+   - Variant definitions: `fastvideo/models/wan/definition.py`
+   - Pipeline config: `fastvideo/models/wan/pipeline_config.py`
+   - Sampling defaults: `fastvideo/pipelines/basic/wan/presets.py`
 
 4. Minimal example.
    - Script: `examples/inference/basic/basic.py`

docs/contributing/performance_benchmarks.md

@@ -23,8 +23,8 @@ It serves three audiences:
 pytest fastvideo/tests/performance/ -vs
 
 # Optional: compare against the rolling HF baseline.
-# PERF_REPORTS_DIR defaults to /root/data/perf_reports for Modal/CI, so
-# override it when running outside the container.
+# PERF_REPORTS_DIR defaults to /root/data/perf_reports in a CI container, so
+# override it for a local run.
 PERF_REPORTS_DIR=/tmp/fastvideo_perf_reports \
 python fastvideo/tests/performance/compare_baseline.py
 
@@ -459,16 +459,18 @@ FlashInfer, Cutlass DSL, SageAttention, Triton, and xFormers when installed.
 | `FASTVIDEO_FA4` | `0` | `test_inference_performance.py` | FlashAttention-4 toggle included in `software_profile_id`. |
 | `FASTVIDEO_PERFORMANCE_PROFILE_VERSION` | unset | `test_inference_performance.py` | Optional explicit software cohort/profile version included in `software_profile_id`. |
 | `IMAGE_VERSION` | unset | `test_inference_performance.py` | CI container image/profile version included in `software_profile_id` when available. |
-| `FASTVIDEO_CONTAINER_IMAGE_REF` | unset | `pr_test.py`, `launch_l40s_job.py`, `test_inference_performance.py` | Resolved CI container image ref or digest recorded in `environment_metadata` for audit without changing `software_profile_id`. |
+| `FASTVIDEO_CONTAINER_IMAGE_REF` | unset | Slurm runner, `test_inference_performance.py` | Pinned CI container image digest recorded in `environment_metadata` and `software_profile_id`. |
 | `FASTVIDEO_STAGE_LOGGING` | set by the pytest test | `test_inference_performance.py` | Enables pipeline stage timing capture for component metrics during benchmark runs. |
 
 ## CI integration
 
-The performance step can run on demand with `/test performance` and as part of
-the Full Suite (see [CI/CD Architecture](ci_architecture.md)). The Modal entry
-point is `fastvideo/tests/modal/pr_test.py:run_performance_tests` and the
-Buildkite artifact upload is in
-`.buildkite/scripts/pr_test.sh:upload_performance_artifacts`.
+The performance step can run on demand with `/test performance`, through a
+merge gate when performance tests or benchmark policy changed, and as part of
+an explicit `/test full` run (see [CI/CD Architecture](ci_architecture.md)).
+The weekly `fastvideo-performance-lane` schedule runs the same Slurm payload.
+The active entry point is `.buildkite/scripts/lanes/performance.sh`; the
+trusted host dispatcher relays its allowlisted reports to Buildkite after the
+isolated container exits.
 
 Each performance build runs pytest first. PR and direct runs only continue to
 `compare_baseline.py` when that fixed-threshold phase passes; if pytest fails,

docs/contributing/pull_requests.md

@@ -42,19 +42,20 @@ The important process labels are:
 
 | Label | Meaning |
 |---|---|
-| `ready` | The PR is ready for Full Suite and auto-merge consideration. |
+| `ready` | The PR is ready for the change-aware merge gate and auto-merge consideration. |
 | `needs-rebase` | The PR has merge conflicts with `main`. |
 | `do-not-merge` | A maintainer has blocked merge. |
 
 ## CI Summary
 
-FastVideo has three validation tiers:
+FastVideo has three routine validation tiers plus an explicit full diagnostic:
 
 | Tier | Runs when | What it does |
 |---|---|---|
 | Pre-commit | Pull requests and `/test pre-commit` | Formatting, linting, typing, spelling, Markdown, workflow syntax, filename checks |
-| Fastcheck | PR Buildkite builds | Path-filtered component and unit checks on Modal GPU runners |
-| Full Suite | `/merge`, `ready`, `/test full`, or new pushes to ready PRs | Path-filtered integration, SSIM, training, eval, API, and performance checks |
+| Fastcheck | PR Buildkite builds | Six component, kernel, unit, and app lanes on Slinky Slurm |
+| Merge gate | `/merge`, `ready`, or new pushes to ready PRs | Only path-relevant integration lanes on Slinky Slurm; Fastcheck remains the baseline |
+| Full Suite | `/test full` | Explicit all-twenty-lane diagnostic run on Slinky Slurm |
 
 See [CI/CD Architecture](ci_architecture.md#ci-tiers) for exact jobs, path
 filters, and workflow files.
@@ -66,10 +67,11 @@ filters, and workflow files.
 3. Fix pre-commit failures locally with `pre-commit run --all-files`.
 4. Wait for at least one approving review.
 5. When the PR is approved and ready, comment `/merge`.
-6. `/merge` adds `ready` and triggers the Full Suite for the PR branch.
+6. `/merge` adds `ready`, waits for cheap checks, and triggers the minimal
+   path-relevant integration lanes for the PR branch.
 7. If all required checks pass, Mergify squash-merges the PR to `main`.
-8. If Full Suite fails, fix the regression, push again, and re-run `/merge` or
-   the failed test.
+8. If the merge gate fails, fix the regression, push again, and re-run
+   `/merge`. Use a targeted `/test` command for diagnosis.
 
 Only contributors with repository write permission can use slash commands. If
 you are an external contributor, ask a maintainer to run `/merge` or add
@@ -128,7 +130,7 @@ git push --force-with-lease
 
 Mergify removes `needs-rebase` after conflicts are resolved.
 
-### Full Suite Fails
+### Merge Gate Or Full Suite Fails
 
 The failing Buildkite step is the source of truth. Common causes are:
 

docs/contributing/testing.md

@@ -10,6 +10,7 @@ slash-command mappings, and workflow ownership live in
 |---|---|---|
 | Unit tests | `fastvideo/tests/api`, `fastvideo/tests/dataset`, `fastvideo/tests/entrypoints`, `fastvideo/tests/workflow`, CPU-safe `fastvideo/tests/train` subsets | Validate individual functions, APIs, contracts, and lightweight workflows. |
 | Component tests | `fastvideo/tests/encoders`, `fastvideo/tests/transformers`, `fastvideo/tests/vaes` | Validate loading and basic behavior for model components. |
+| Golden gates | `fastvideo/tests/golden_gate` | Compare small, deterministic component outputs exactly against device/runtime-matched reference tensors. |
 | Train framework tests | `fastvideo/tests/train/models`, `fastvideo/tests/train/methods` | Exercise the new `fastvideo/train/` framework on real checkpoints and tiny synthetic batches. |
 | SSIM tests | `fastvideo/tests/ssim` | Compare generated videos against references to catch visual regressions. |
 | Training tests | `fastvideo/tests/training` | Validate legacy training loops, LoRA, distillation, self-forcing, and VSA behavior. |
@@ -20,14 +21,74 @@ slash-command mappings, and workflow ownership live in
 
 ## Running Tests Locally
 
-Run the narrowest useful suite while iterating:
+Start with the cheapest checks that cover the changed behavior, and stop on a
+failure before starting heavier dependent checks:
+
+1. Run pre-commit on changed files and focused import, config, and contract tests.
+2. Run the smallest matching component golden gate. Prefer one GPU, tiny fixed
+   inputs, cached component weights, and direct tensor comparisons over renders.
+3. Run focused component parity or default SSIM when the golden does not cover
+   the changed behavior, such as VAE normalization or pipeline wiring.
+4. Run full-quality renders or broad suites when required by the change, an
+   explicit request, or CI policy, rather than on every edit.
+
+A golden must cover the component being changed. Wan has four small gates:
+
+| Gate | Boundary |
+|---|---|
+| `test_wan_t2v.py` | Dense transformer block 0 |
+| `test_wan_vae.py` | FP32 encode, BF16 decode, streaming, and cache reset |
+| `test_wan_causal.py` | Real block weights, cache append/rewrite, and sink eviction |
+| `test_wan_denoising.py` | Three real 1.3B DiT/UniPC steps with fixed prompt embeddings |
+
+These use an immutable Wan checkpoint revision and only download the required
+component. The trajectory gate needs no tokenizer, text encoder, VAE, or video
+reference. Weight-free stage tests also exercise every step of a 50-step UniPC
+loop, CFG caching, expert switching, conditioning layouts, and DMD RNG order.
+These checks do not replace independent Diffusers parity or end-to-end SSIM.
+
+Use the golden's matching GPU, dtype, backend, and runtime. A missing reference
+or environment mismatch is not a pass. Do not replace a reference with the
+candidate output just to clear a failure. For a relocation, the unchanged
+parent is the baseline; two imports of the same class are not numerical proof.
+
+The VAE and transformer CI lanes run their Wan component goldens first.
+Selected integration lanes wait for the golden lane in merge/full builds.
+See [CI/CD Architecture](ci_architecture.md) for direct-rerun and skip semantics.
+
+For Wan, one command enforces the local ordering and stops on failure:
+
+The initial contracts include `tests/api/test_wan_definitions.py`: all registered
+Wan aliases, local-manifest detector precedence, sampling/precision defaults,
+config isolation, and legacy serialized-config compatibility. These require no
+weights or Hub access; the current package import still needs its prepared
+runtime environment.
 
 ```bash
-pytest tests/
-pytest fastvideo/tests/ -v
-pytest fastvideo/tests/encoders -vs
-pytest fastvideo/tests/transformers -vs
-pytest fastvideo/tests/vaes -vs
+bash scripts/validate_wan.sh vae           # contracts, then the VAE golden
+bash scripts/validate_wan.sh dense parity  # contracts, goldens, Diffusers parity
+bash scripts/validate_wan.sh all default   # then focused T2V/I2V/causal SSIM
+```
+
+The second argument is an upper validation tier, not a reference override.
+Supply the GPU/backend/runtime and SSIM model/tier settings that match the
+reference. The script never updates references. Causal component coverage is
+a block-cache fingerprint, not independent full causal-pipeline parity.
+
+New named-tensor gates write a missing output to `*.candidate.pt` and fail;
+running candidate code again cannot turn it into an approved reference. Seed
+from unchanged, pushed source, verify two independent processes bit-for-bit,
+then review and publish only the new reference files. Preserve the baseline
+source SHA, test recipe SHA, checkpoint revision, runtime, and comparison
+receipt with the run artifacts. Never overwrite an existing video reference
+as a side effect of adding a tensor gate.
+
+Examples of focused checks:
+
+```bash
+pytest fastvideo/tests/loader/test_wan_family_imports.py -q
+pytest fastvideo/tests/golden_gate/test_wan_t2v.py -q
+pytest fastvideo/tests/vaes/test_wan_vae.py -q
 ```
 
 GPU-heavy suites need the right hardware, credentials, local caches, and
@@ -138,47 +199,31 @@ pytest fastvideo/tests/ssim/ -vs
 
 Use a machine whose GPU and backend match the reference folder you are testing.
 
-## Modal Runs For SSIM
+## Slurm CI Runs For SSIM
 
-For CI-like SSIM execution, use `fastvideo/tests/modal/ssim_test.py`:
+Comment `/test ssim` on a pull request to run the canonical four-GPU SSIM
+lane on the Slinky Slurm cluster. `fastvideo/tests/ssim/ci_runner.py`
+discovers the suite without importing test modules, packs independent pytest
+processes across the four assigned GPUs, and stops the lane on the first
+failure.
+
+The change-aware `/merge` planner may run only the SSIM test basenames owned
+by the changed model family. Shared SSIM harness changes still select the
+complete lane. Independently, `main` runs the full SSIM matrix every Sunday at
+05:00 UTC so infrequently touched model families retain periodic coverage.
+
+For a focused developer run, invoke pytest directly and optionally select one
+model from a parameterized test through `FASTVIDEO_SSIM_MODEL_ID`:
 
 ```bash
-python -m modal run fastvideo/tests/modal/ssim_test.py::run_ssim_tests
+pytest fastvideo/tests/ssim/test_wan_t2v_similarity.py -vs
+
+FASTVIDEO_SSIM_MODEL_ID=Wan2.1-T2V-1.3B-Diffusers \
+pytest fastvideo/tests/ssim/test_wan_t2v_similarity.py -vs
 ```
 
-Target specific files or model ids:
-
-```bash
-python -m modal run fastvideo/tests/modal/ssim_test.py::run_ssim_tests \
-  --test-files test_wan_t2v_similarity.py \
-  --model-ids Wan2.1-T2V-1.3B-Diffusers
-```
-
-If `HF_API_KEY`, `HUGGINGFACE_HUB_TOKEN`, or `HF_TOKEN` is not set, the local
-entrypoint fails fast.
-
-To export raw generated videos from Modal to the shared volume:
-
-```bash
-python -m modal run fastvideo/tests/modal/ssim_test.py::run_ssim_tests \
-  --sync-generated-to-volume
-```
-
-The raw export path is quality-tiered:
-
-- default params: `ssim_generated_videos/default/<subdir>/generated_videos`
-- full-quality params: `ssim_generated_videos/full_quality/<subdir>/generated_videos`
-
-The printed `modal volume get` command downloads into
-`./generated_videos_modal/<quality-tier>`. Convert those outputs into local
-references with `copy-local`:
-
-```bash
-python fastvideo/tests/ssim/reference_videos_cli.py copy-local \
-  --quality-tier full_quality \
-  --generated-dir ./generated_videos_modal/full_quality/L40S_reference_videos \
-  --device-folder L40S_reference_videos
-```
+The files under `fastvideo/tests/modal/` are retained only as a disabled
+manual rollback implementation. No active CI trigger invokes them.
 
 ### SSIM Bootstrap Mode
 
@@ -206,28 +251,30 @@ python fastvideo/tests/ssim/reference_videos_cli.py promote-draft \
 
 ## CI Integration
 
-FastVideo CI tests are orchestrated by Buildkite and run on Modal GPU
-instances. The main files are:
+FastVideo GPU CI is orchestrated by Buildkite and runs only on isolated Slinky
+Slurm workers. The main files are:
 
 | File | Purpose |
 |---|---|
-| `.buildkite/pipeline.yml` | Buildkite test graph and path filters. |
-| `.buildkite/scripts/pr_test.sh` | Dispatches `TEST_TYPE` to a Modal function. |
-| `fastvideo/tests/modal/pr_test.py` | Modal functions for most test lanes. |
-| `fastvideo/tests/modal/ssim_test.py` | Modal functions and partitioning for SSIM. |
+| `.buildkite/pipeline.yml` | Static, validated 20-lane Slurm test graph. |
+| `.github/scripts/plan_merge_ci.py` | Trusted path-to-lane and focused quality-test policy for `/merge`. |
+| `.buildkite/scripts/unit_test.sh`, `.buildkite/scripts/lanes/*.sh` | Repository-owned test payloads executed inside Slurm containers. |
+| `fastvideo/tests/ssim/ci_runner.py` | Four-GPU SSIM task discovery and scheduling. |
+| `.buildkite/scripts/pr_test.sh`, `fastvideo/tests/modal/*.py` | Dormant manual rollback path; rejected in Buildkite. |
 
-For exact tier membership, path filters, slash commands, and aggregate statuses,
+For exact tier membership, slash commands, runner isolation, and aggregate statuses,
 see [CI/CD Architecture](ci_architecture.md).
 
 ### Adding A New CI Test Category
 
 If a new test does not fit an existing lane:
 
-1. Add a Modal function in `fastvideo/tests/modal/pr_test.py` or a focused
-   companion module.
-2. Add a matching `TEST_TYPE` case in `.buildkite/scripts/pr_test.sh`.
-3. Add Buildkite direct-test and path-filter entries in `.buildkite/pipeline.yml`.
-4. Add the `/test` mapping in `.github/workflows/ci-slash-commands.yml`.
+1. Put the test payload in an executable `.buildkite/scripts/lanes/<lane>.sh`.
+2. Add its static step to `.buildkite/pipeline.yml`, its changed-path ownership
+   to `.github/scripts/plan_merge_ci.py`, and extend the CI contract tests.
+3. Add the `/test` mapping in `.github/workflows/ci-slash-commands.yml`.
+4. Coordinate the matching GPU, timeout, dependency, secret, and artifact
+   policy in the private Slurm runner allowlist.
 5. Document the new category in [CI/CD Architecture](ci_architecture.md) and add
    authoring notes here if contributors need them.
 

docs/cookbook/cosmos.md

@@ -0,0 +1,140 @@
+---
+hide:
+- toc
+---
+
+# Cosmos recipes
+
+<div class="cookbook-shell cookbook-family-page" data-cookbook data-family="cosmos" data-recipes="../../assets/cookbook-recipes.json?v=8">
+  <header class="cookbook-family-header">
+    <a class="cookbook-back-link" href="../"><span aria-hidden="true">←</span> All model families</a>
+    <div class="cookbook-family-header__body">
+      <span class="cookbook-family-header__logo">
+        <img class="off-glb" src="../../assets/logos/nvidia.webp" alt="NVIDIA" width="112" height="112">
+      </span>
+      <div>
+        <p class="cookbook-eyebrow">Maintained family · Inference</p>
+        <h2>Cosmos inference recipes</h2>
+        <p>NVIDIA Cosmos Predict 2.5 generates navigable world videos. The maintained example runs the 2B text-to-world checkpoint on a single GPU.</p>
+      </div>
+    </div>
+    <div class="cookbook-lifecycle" aria-label="Lifecycle stages">
+      <span class="cookbook-lifecycle__stage cookbook-lifecycle__stage--active">Inference <small>live</small></span>
+      <span class="cookbook-lifecycle__stage">Distillation <small>planned</small></span>
+      <span class="cookbook-lifecycle__stage">Fine-tuning <small>planned</small></span>
+      <span class="cookbook-lifecycle__stage">Training <small>planned</small></span>
+      <span class="cookbook-lifecycle__stage">Evaluation <small>planned</small></span>
+      <span class="cookbook-lifecycle__stage">Optimization <small>planned</small></span>
+      <span class="cookbook-lifecycle__stage">Deployment <small>planned</small></span>
+    </div>
+  </header>
+  <nav class="cookbook-jumpnav" aria-label="Recipe page sections">
+    <a href="#recipe-builder">Builder</a>
+    <a href="#cookbook-setup">Setup</a>
+    <a href="#cookbook-troubleshooting">Troubleshooting</a>
+    <a href="#cookbook-evidence">Evidence</a>
+  </nav>
+
+  <section class="cookbook-builder" id="recipe-builder" aria-labelledby="builder-heading">
+    <div class="cookbook-builder__intro">
+      <h2 id="builder-heading">Pick a recipe and runtime</h2>
+      <p>Start with the result you want, then choose one of the runtimes FastVideo actually maintains for it.</p>
+    </div>
+
+    <div class="cookbook-builder__layout">
+      <div class="cookbook-controls">
+        <div class="cookbook-selection-row">
+          <div class="cookbook-selection-row__label">
+            <strong>Recipe</strong>
+            <span>Task and checkpoint</span>
+          </div>
+          <div class="cookbook-option-grid cookbook-option-grid--models" data-cookbook-model-options role="group" aria-label="Recipe">
+            <button type="button" disabled>Loading Cosmos recipes...</button>
+          </div>
+        </div>
+
+        <div class="cookbook-selection-row">
+          <div class="cookbook-selection-row__label">
+            <strong>Runtime</strong>
+            <span>Maintained paths only</span>
+          </div>
+          <div class="cookbook-option-grid cookbook-option-grid--hardware" data-cookbook-hardware-options role="group" aria-label="Runtime">
+            <button type="button" disabled>Loading runtimes...</button>
+          </div>
+        </div>
+
+        <p class="cookbook-selection-description" data-cookbook-description>Loading recipe details...</p>
+        <p class="cookbook-hardware-note">Exact device and memory details appear only when a recorded run supports them.</p>
+
+        <div class="cookbook-hardware-state" data-cookbook-hardware-state role="status" aria-live="polite">
+          Reading recipe evidence...
+        </div>
+      </div>
+
+      <article class="cookbook-result">
+        <div class="cookbook-result__header">
+          <h3 data-cookbook-label>Loading...</h3>
+          <div class="cookbook-result__badges">
+            <span class="cookbook-badge">Maintained</span>
+            <span class="cookbook-badge" data-cookbook-evidence>Source-backed</span>
+            <span class="cookbook-badge cookbook-badge--neutral" data-cookbook-hardware-badge>Source config</span>
+          </div>
+        </div>
+
+        <dl class="cookbook-result__facts">
+          <div><dt>Model</dt><dd data-cookbook-model>Loading...</dd></div>
+          <div><dt>Workload</dt><dd data-cookbook-task>Loading...</dd></div>
+          <div><dt>Source configuration</dt><dd data-cookbook-gpus>Loading...</dd></div>
+          <div><dt>Expected output</dt><dd data-cookbook-artifact>Loading...</dd></div>
+        </dl>
+
+        <div class="cookbook-command">
+          <div class="cookbook-command__bar">
+            <span>Terminal</span>
+          </div>
+          <pre><code class="language-bash" data-cookbook-command>Loading...</code></pre>
+        </div>
+
+        <div class="cookbook-result__footer">
+          <a data-cookbook-source href="../../inference/examples/basic/">Open example source</a>
+          <a data-cookbook-model-link href="https://huggingface.co/nvidia">View model card</a>
+        </div>
+        <p class="cookbook-picker__status" role="status" aria-live="polite" data-cookbook-status></p>
+      </article>
+    </div>
+
+    <noscript>
+      <div class="cookbook-noscript">
+        JavaScript is needed for the guided selector. You can still browse the
+        <a href="../../inference/examples/examples_inference_index/">maintained inference examples</a>.
+      </div>
+    </noscript>
+  </section>
+</div>
+
+<details class="cookbook-collapsible" id="cookbook-setup">
+  <summary>Setup</summary>
+  <div class="cookbook-collapsible__body">
+      <p>The generated commands expect a local clone:</p>
+      <pre><code>git clone https://github.com/hao-ai-lab/FastVideo.git
+cd FastVideo</code></pre>
+      <p>Use <a href="../../inference/configuration/">Configuration</a> for supported Python and CLI settings, <a href="../../inference/optimizations/">Optimizations</a> for attention and memory tradeoffs, and the <a href="../../inference/support_matrix/">support matrix</a> for the supported model and optimization surface.</p>
+  </div>
+</details>
+
+<details class="cookbook-collapsible" id="cookbook-troubleshooting">
+  <summary>Troubleshooting</summary>
+  <div class="cookbook-collapsible__body">
+      <ul>
+        <li>World-generation prompts work best describing a scene and camera motion; the built-in prompt in the example is a known-good starting point.</li>
+        <li>Gated or missing checkpoints: run <code>huggingface-cli login</code> and confirm you accepted the model's license on Hugging Face.</li>
+      </ul>
+  </div>
+</details>
+
+<details class="cookbook-collapsible" id="cookbook-evidence">
+  <summary>Evidence status</summary>
+  <div class="cookbook-collapsible__body">
+      <p>All recipes on this page are <strong>Source-backed</strong>: their commands, model IDs, and flags were validated against the checked-in FastVideo sources listed above (static validation). No runtime GPU validation is recorded for these recipes, so GPU model fit, memory use, throughput, and runtime duration are <strong>Unknown</strong> and deliberately not claimed. Runtime buttons show only the GPU counts configured in checked-in sources.</p>
+  </div>
+</details>