ckpt

launched synthetic video generation for hy1.5
ckpt
2026-04-01 02:22:08 +00:00 · 2026-02-08 23:46:16 +00:00 · 2026-02-07 21:43:09 +00:00 · 2026-01-28 19:10:28 +00:00 · 2026-01-28 15:39:18 +00:00 · 2026-01-28 15:32:15 +00:00
305 changed files with 10071 additions and 18693 deletions
@@ -67,9 +67,6 @@ jobs:
            - torch-version: '2.9.1'
              cuda-version: '12.8.0'
              torch-cuda-short: 'cu128'
-            # - torch-version: '2.10.0'
-            #   cuda-version: '12.8.0'
-            #   torch-cuda-short: 'cu128'

    steps:
      - name: Free up disk space
@@ -223,6 +220,3 @@ jobs:
        uses: pypa/gh-action-pypi-publish@release/v1
        with:
          packages-dir: fastvideo-kernel/dist/
-          # PyPI does not allow replacing an existing file with the same name.
-          # This makes re-runs idempotent by skipping files already uploaded.
-          skip-existing: true
@@ -18,7 +18,6 @@ venv/
 .venv/
 runs/
 samples/
-Miniconda3-latest-Linux-x86_64.sh
 *validation/
 data/
 outputs/
@@ -33,11 +32,6 @@ env
 **.txt
 *.log
 weights/
-official_weights/
-converted_weights/
-
-# SSIM test outputs
-fastvideo/tests/ssim/generated_videos/

 # Distribution / packaging
 build/
@@ -75,11 +69,6 @@ docs/distillation/examples/
 !docs/assets/images/**/*.png
 !comfyui/assets/**/*.png
 !comfyui/assets/**/*.gif
-!assets/images/**/*.png
-!assets/images/**/*.jpg
-!assets/images/**/*.jpeg
-!assets/images/**/*.gif
-!assets/videos/**/*.mp4

 dmd_t2v_output/
 preprocess_output_text/
@@ -10,7 +10,7 @@ exclude: |
    demo/.*|
    predict\.py|
    scripts/.*|
-    assets/prompts/.*|
+    prompts/.*|
    fastvideo/data_preprocess/.*|
    fastvideo/dataset/.*|
    fastvideo/models/.*|
@@ -1,42 +0,0 @@
-# Repository Guidelines
-
-## Project Structure & Module Organization
- Core Python package: `fastvideo/` (models, pipelines, training, distributed runtime, CLI entrypoints).
- CUDA/custom kernels: `fastvideo-kernel/` (separate build/test flow).
- Tests:
-  - `fastvideo/tests/` for package-level tests (dataset, encoders, inference, training, SSIM, workflow).
-  - `tests/local_tests/` for additional local/component checks.
- Docs and guides: `docs/` (MkDocs source), with contributor docs in `docs/contributing/`.
- Runnable examples and scripts: `examples/` and `scripts/`.
- Static assets: `assets/` (including `assets/images/`, `assets/videos/`, and `assets/prompts/`) and `comfyui/assets/`.
-
-## Build, Test, and Development Commands
- `uv pip install -e .[dev]`: editable install with lint/test extras.
- `pre-commit install --hook-type pre-commit --hook-type commit-msg`: enable local hooks.
- `pre-commit run --all-files`: run formatter/lint/type/spelling checks.
- `pytest tests/`: run top-level test suite.
- `pytest fastvideo/tests/ -v`: run package tests.
- `pytest fastvideo/tests/ssim/ -vs`: run SSIM regression tests (GPU-heavy).
- `cd fastvideo-kernel && ./build.sh`: build kernel extensions.
-
-## Coding Style & Naming Conventions
- Python 3.10+; 4-space indentation; keep code and imports readable and explicit.
- Style tools are configured in `pyproject.toml` and `.pre-commit-config.yaml`:
-  - `yapf` (format), `ruff` (lint, auto-fix), `mypy` (typing), `codespell`.
- Target line length is 80.
- Naming: `snake_case` for functions/files, `PascalCase` for classes, `UPPER_SNAKE_CASE` for constants.
-
-## Testing Guidelines
- Use `pytest` and place tests near relevant domains (e.g., `fastvideo/tests/encoders/`).
- Prefer descriptive names like `test_<feature>_<expected_behavior>.py`.
- For new pipelines/backends, include at least one regression-oriented test; add SSIM coverage when output quality must be preserved.
- Document GPU assumptions in tests that require specific hardware.
-
-## Commit & Pull Request Guidelines
- Follow existing commit style: short subject with optional tag prefix, e.g. `[bugfix]: ...`, `[feat]: ...`, `[misc]: ...`, and include PR reference like `(#1234)` when applicable.
- Keep commits focused by concern (feature, refactor, fix).
- PRs should include:
-  - clear problem/solution summary,
-  - test evidence (`pytest`/SSIM outputs or rationale if skipped),
-  - linked issue/PR context,
-  - screenshots or sample outputs for UI/demo/docs changes.
@@ -1 +0,0 @@
-@AGENTS.md
@@ -3,32 +3,33 @@
 </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://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://ibb.co/sv3MMKyv" target="_blank"> <b> WeChat </b> </a> |
 </p>

 **FastVideo is a unified post-training and inference framework for accelerated video generation.**

 ## NEWS
+- ```2025/11/19```: Release [CausalWan2.2 I2V A14B Preview](https://huggingface.co/FastVideo/CausalWan2.2-I2V-A14B-Preview-Diffusers) models, [Blog](https://hao-ai-lab.github.io/blogs/fastvideo_causalwan_preview/) and [Inference Code!](https://github.com/hao-ai-lab/FastVideo/blob/main/examples/inference/basic/basic_self_forcing_causal_wan2_2_i2v.py)
+- ```2025/08/04```: Release [FastWan](https://hao-ai-lab.github.io/FastVideo/distillation/dmd) models and [Sparse-Distillation](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/).

- `2025/11/19`: Release [CausalWan2.2 I2V A14B Preview](https://huggingface.co/FastVideo/CausalWan2.2-I2V-A14B-Preview-Diffusers) models, [Blog](https://hao-ai-lab.github.io/blogs/fastvideo_causalwan_preview/) and [Inference Code!](https://github.com/hao-ai-lab/FastVideo/blob/main/examples/inference/basic/basic_self_forcing_causal_wan2_2_i2v.py)
- `2025/08/04`: Release [FastWan](https://hao-ai-lab.github.io/FastVideo/distillation/dmd) models and [Sparse-Distillation](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/).
+<details>
+<summary>More</summary>

-### More News
+- ```2025/06/14```: Release finetuning and inference code for [VSA](https://arxiv.org/pdf/2505.13389)
+- ```2025/04/24```: [FastVideo V1](https://hao-ai-lab.github.io/blogs/fastvideo/) is released!
+- ```2025/02/18```: Release the inference code for [Sliding Tile Attention](https://hao-ai-lab.github.io/blogs/sta/).

- `2025/06/14`: Release finetuning and inference code for [VSA](https://arxiv.org/pdf/2505.13389)
- `2025/04/24`: [FastVideo V1](https://hao-ai-lab.github.io/blogs/fastvideo/) is released!
- `2025/02/18`: Release the inference code for [Sliding Tile Attention](https://hao-ai-lab.github.io/blogs/sta/).
+</details>

 ## Key Features

 FastVideo has the following features:
-
 - End-to-end post-training support for bidirectional and autoregressive models:
  - Support full finetuning and LoRA finetuning for state-of-the-art open video DiTs
  - Data preprocessing pipeline for video, image, and text data
  - Distribution Matching Distillation (DMD2) stepwise distillation.
  - Sparse attention with [Video Sparse Attention](https://arxiv.org/pdf/2505.13389)
-  - [Sparse distillation](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/) to achieve >50x denoising speedup
+  - [Sparse distillation](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/) to achineve >50x denoising speedup
  - Scalable training with FSDP2, sequence parallelism, and selective activation checkpointing.
  - Causal distillation through Self-Forcing
  - See this [page](https://hao-ai-lab.github.io/FastVideo/training/overview/) for full list of supported models and recipes.
@@ -43,7 +44,6 @@ FastVideo has the following features:
  - See this [page](https://hao-ai-lab.github.io/FastVideo/inference/hardware_support/) for full list of supported hardware and OS.

 ## Getting Started
-
 We recommend using an environment manager such as `Conda` to create a clean environment:

 ```bash
@@ -58,20 +58,18 @@ pip install fastvideo
 Please see our [docs](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/) for more detailed installation instructions.

 ## Sparse Distillation
-
 For our sparse distillation techniques, please see our [distillation docs](https://hao-ai-lab.github.io/FastVideo/distillation/dmd/) and check out our [blog](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/).

 See below for recipes and datasets:

-| Model                                                                                 | Sparse Distillation                                                                                             | Dataset                                                                                                  |
-| ------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------- |
-| [FastWan2.1-T2V-1.3B](https://huggingface.co/FastVideo/FastWan2.1-T2V-1.3B-Diffusers) | [Recipe](https://github.com/hao-ai-lab/FastVideo/tree/main/examples/distill/Wan2.1-T2V/Wan-Syn-Data-480P)       | [FastVideo Synthetic Wan2.1 480P](https://huggingface.co/datasets/FastVideo/Wan-Syn_77x448x832_600k)     |
-| [FastWan2.2-TI2V-5B](https://huggingface.co/FastVideo/FastWan2.2-TI2V-5B-Diffusers)   | [Recipe](https://github.com/hao-ai-lab/FastVideo/tree/main/examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free) | [FastVideo Synthetic Wan2.2 720P](https://huggingface.co/datasets/FastVideo/Wan2.2-Syn-121x704x1280_32k) |
+|                                            Model                                              |                                               Sparse Distillation                                                 |                                                  Dataset                                                  |
+|:-------------------------------------------------------------------------------------------:  |:---------------------------------------------------------------------------------------------------------------:  |:--------------------------------------------------------------------------------------------------------: |
+| [FastWan2.1-T2V-1.3B](https://huggingface.co/FastVideo/FastWan2.1-T2V-1.3B-Diffusers)         |    [Recipe](https://github.com/hao-ai-lab/FastVideo/tree/main/examples/distill/Wan2.1-T2V/Wan-Syn-Data-480P)      | [FastVideo Synthetic Wan2.1 480P](https://huggingface.co/datasets/FastVideo/Wan-Syn_77x448x832_600k)      |
+| [FastWan2.1-T2V-14B-Preview](https://huggingface.co/FastVideo/FastWan2.1-T2V-14B-Diffusers)   |                                                   Coming soon!                                                    |   [FastVideo Synthetic Wan2.1 720P](https://huggingface.co/datasets/FastVideo/Wan-Syn_77x768x1280_250k)   |
+| [FastWan2.2-TI2V-5B](https://huggingface.co/FastVideo/FastWan2.2-TI2V-5B-Diffusers)           | [Recipe](https://github.com/hao-ai-lab/FastVideo/tree/main/examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free)   | [FastVideo Synthetic Wan2.2 720P](https://huggingface.co/datasets/FastVideo/Wan2.2-Syn-121x704x1280_32k)  |

 ## Inference
-
 ### Generating Your First Video
-
 Here's a minimal example to generate a video using the default settings. Make sure VSA kernels are [installed](https://hao-ai-lab.github.io/FastVideo/video_sparse_attention/installation/). Create a file called `example.py` with the following code:

 ```python
@@ -110,36 +108,55 @@ python example.py

 For a more detailed guide, please see our [inference quick start](https://hao-ai-lab.github.io/FastVideo/inference/inference_quick_start/).

-## More Guides
+### Other docs:

 - [Design Overview](https://hao-ai-lab.github.io/FastVideo/design/overview/)
+- [Contribution Guide](https://hao-ai-lab.github.io/FastVideo/getting_started/installation/)
+
+## Distillation and Finetuning
 - [Distillation Guide](https://hao-ai-lab.github.io/FastVideo/distillation/dmd/)
- [Contribution Guide](https://hao-ai-lab.github.io/FastVideo/contributing/overview/)
+<!-- - [Finetuning Guide](https://hao-ai-lab.github.io/FastVideo/training/finetune.html) -->

 ## Awesome work using FastVideo or our research projects

- [SGLang](https://github.com/sgl-project/sglang/tree/main/python/sglang/multimodal_gen): SGLang's diffusion inference functionality is based on a fork of FastVideo on Sept. 24, 2025.
- [DanceGRPO](https://github.com/XueZeyue/DanceGRPO): A unified framework to adapt Group Relative Policy Optimization (GRPO) to visual generation paradigms. Code based on FastVideo.
- [SRPO](https://github.com/Tencent-Hunyuan/SRPO): A method to directly align the full diffusion trajectory with fine-grained human preference. Code based on FastVideo.
- [DCM](https://github.com/Vchitect/DCM): Dual-expert consistency model for efficient and high-quality video generation. Code based on FastVideo.
- [Hunyuan Video 1.5](https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5): A leading lightweight video generation model, where they proposed SSTA based on Sliding Tile Attention.
- [Kandinsky-5.0](https://github.com/kandinskylab/kandinsky-5): A family of diffusion models for video & image generation, where their NABLA attention includes a Sliding Tile Attention branch.
- [LongCat Video](https://github.com/meituan-longcat/LongCat-Video): A foundational video generation model with 13.6B parameters with block-sparse attention similar to Video Sparse Attention.
+- [SGLang](https://github.com/sgl-project/sglang/tree/main/python/sglang/multimodal_gen): SGLang's diffusion inference functionality is based  on a fork of FastVideo on Sept. 24, 2025. [![Star](https://img.shields.io/github/stars/sgl-project/sglang.svg?style=social&label=Star)](https://github.com/sgl-project/sglang)
+
+- [DanceGRPO](https://github.com/XueZeyue/DanceGRPO): A  unified framework to adapt Group Relative Policy Optimization (GRPO) to visual generation paradigms. Code based on FastVideo. [![Star](https://img.shields.io/github/stars/XueZeyue/DanceGRPO.svg?style=social&label=Star)](https://github.com/XueZeyue/DanceGRPO)
+- [SRPO](https://github.com/Tencent-Hunyuan/SRPO): A method to directly align the full diffusion trajectory with fine-grained human preference. Code based on FastVideo. [![Star](https://img.shields.io/github/stars/Tencent-Hunyuan/SRPO.svg?style=social&label=Star)](https://github.com/Tencent-Hunyuan/SRPO)
+- [DCM](https://github.com/Vchitect/DCM): Dual-expert consistency model for efficient and high-quality video generation. Code based on FastVideo. [![Star](https://img.shields.io/github/stars/Vchitect/DCM.svg?style=social&label=Star)](https://github.com/Vchitect/DCM)
+- [Hunyuan Video 1.5](https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5): A leading lightweight video generation model, where they proposed SSTA based on Sliding Tile Attention. [![Star](https://img.shields.io/github/stars/Tencent-Hunyuan/HunyuanVideo-1.5.svg?style=social&label=Star)](https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5)
+- [Kandinsky-5.0](https://github.com/kandinskylab/kandinsky-5): A family of diffusion models for video & image generation, where their NABLA attention includes a Sliding Tile Attention branch. [![Star](https://img.shields.io/github/stars/kandinskylab/kandinsky-5.svg?style=social&label=Star)](https://github.com/kandinskylab/kandinsky-5)
+- [LongCat Video](https://github.com/meituan-longcat/LongCat-Video): A foundational video generation model with 13.6B parameters with block-sparse attention similar to Video Sparse Attention. [![Star](https://img.shields.io/github/stars/meituan-longcat/LongCat-Video.svg?style=social&label=Star)](https://github.com/meituan-longcat/LongCat-Video)

 ## 🤝 Contributing

 We welcome all contributions. Please check out our guide [here](https://hao-ai-lab.github.io/FastVideo/contributing/overview/).
 See details in [development roadmap](https://github.com/hao-ai-lab/FastVideo/issues/899).
-
 ## Acknowledgement
+We learned and reused code from the following projects:
+- [Wan-Video](https://github.com/Wan-Video)
+- [ThunderKittens](https://github.com/HazyResearch/ThunderKittens)
+- [Triton](https://github.com/triton-lang/triton)
+- [DMD2](https://github.com/tianweiy/DMD2)
+- [diffusers](https://github.com/huggingface/diffusers)
+- [xDiT](https://github.com/xdit-project/xDiT)
+- [vLLM](https://github.com/vllm-project/vllm)
+- [SGLang](https://github.com/sgl-project/sglang)

-We learned the design and reused code from the following projects: [Wan-Video](https://github.com/Wan-Video), [ThunderKittens](https://github.com/HazyResearch/ThunderKittens), [DMD2](https://github.com/tianweiy/DMD2), [diffusers](https://github.com/huggingface/diffusers), [xDiT](https://github.com/xdit-project/xDiT), [vLLM](https://github.com/vllm-project/vllm), [SGLang](https://github.com/sgl-project/sglang). We thank [MBZUAI](https://ifm.mbzuai.ac.ae/), [Anyscale](https://www.anyscale.com/), and [GMI Cloud](https://www.gmicloud.ai/) for their support throughout this project.
+We thank [MBZUAI](https://ifm.mbzuai.ac.ae/), [Anyscale](https://www.anyscale.com/), and [GMI Cloud](https://www.gmicloud.ai/) for their support throughout this project.

 ## Citation
-
-If you find FastVideo useful, please consider citing our research work:
+If you find FastVideo useful, please considering citing our work:

 ```bibtex
+@software{fastvideo2024,
+  title        = {FastVideo: A Unified Framework for Accelerated Video Generation},
+  author       = {The FastVideo Team},
+  url          = {https://github.com/hao-ai-lab/FastVideo},
+  month        = apr,
+  year         = {2024},
+}
+
@article{zhang2025vsa,
  title={Vsa: Faster video diffusion with trainable sparse attention},
  author={Zhang, Peiyuan and Chen, Yongqi and Huang, Haofeng and Lin, Will and Liu, Zhengzhong and Stoica, Ion and Xing, Eric and Zhang, Hao},
@@ -1,7 +0,0 @@
-# FastVideo/assets/videos
-
-This folder is used to store **video assets for examples**, primarily **input videos** consumed by scripts under `FastVideo/examples/`.
-
- **Typical contents**: short input clips for demos (e.g., video2world / image2video examples).
- **Non-critical**: these assets are for convenience and are not required to use the FastVideo library.
- **Large files**: avoid committing large videos to git; prefer shared storage or download-on-demand.
@@ -0,0 +1,195 @@
+import argparse
+import os
+import tempfile
+
+import gradio as gr
+import torch
+from diffusers import FlowMatchEulerDiscreteScheduler
+from diffusers.utils import export_to_video
+
+from fastvideo.distill.solver import PCMFMScheduler
+from fastvideo.models.mochi_hf.modeling_mochi import MochiTransformer3DModel
+from fastvideo.models.mochi_hf.pipeline_mochi import MochiPipeline
+
+
+def init_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--prompts", nargs="+", default=[])
+    parser.add_argument("--num_frames", type=int, default=25)
+    parser.add_argument("--height", type=int, default=480)
+    parser.add_argument("--width", type=int, default=848)
+    parser.add_argument("--num_inference_steps", type=int, default=8)
+    parser.add_argument("--guidance_scale", type=float, default=4.5)
+    parser.add_argument("--model_path", type=str, default="data/mochi")
+    parser.add_argument("--seed", type=int, default=12345)
+    parser.add_argument("--transformer_path", type=str, default=None)
+    parser.add_argument("--scheduler_type", type=str, default="pcm_linear_quadratic")
+    parser.add_argument("--lora_checkpoint_dir", type=str, default=None)
+    parser.add_argument("--shift", type=float, default=8.0)
+    parser.add_argument("--num_euler_timesteps", type=int, default=50)
+    parser.add_argument("--linear_threshold", type=float, default=0.1)
+    parser.add_argument("--linear_range", type=float, default=0.75)
+    parser.add_argument("--cpu_offload", action="store_true")
+    return parser.parse_args()
+
+
+def load_model(args):
+    if args.scheduler_type == "euler":
+        scheduler = FlowMatchEulerDiscreteScheduler()
+    else:
+        linear_quadratic = True if "linear_quadratic" in args.scheduler_type else False
+        scheduler = PCMFMScheduler(
+            1000,
+            args.shift,
+            args.num_euler_timesteps,
+            linear_quadratic,
+            args.linear_threshold,
+            args.linear_range,
+        )
+
+    if args.transformer_path:
+        transformer = MochiTransformer3DModel.from_pretrained(args.transformer_path)
+    else:
+        transformer = MochiTransformer3DModel.from_pretrained(args.model_path, subfolder="transformer/")
+
+    pipe = MochiPipeline.from_pretrained(args.model_path, transformer=transformer, scheduler=scheduler)
+    pipe.enable_vae_tiling()
+    # pipe.to(device)
+    # if args.cpu_offload:
+    pipe.enable_sequential_cpu_offload()
+    return pipe
+
+
+def generate_video(
+    prompt,
+    negative_prompt,
+    use_negative_prompt,
+    seed,
+    guidance_scale,
+    num_frames,
+    height,
+    width,
+    num_inference_steps,
+    randomize_seed=False,
+):
+    if randomize_seed:
+        seed = torch.randint(0, 1000000, (1, )).item()
+
+    generator = torch.Generator(device="cuda").manual_seed(seed)
+
+    if not use_negative_prompt:
+        negative_prompt = None
+
+    with torch.autocast("cuda", dtype=torch.bfloat16):
+        output = pipe(
+            prompt=[prompt],
+            negative_prompt=negative_prompt,
+            height=height,
+            width=width,
+            num_frames=num_frames,
+            num_inference_steps=num_inference_steps,
+            guidance_scale=guidance_scale,
+            generator=generator,
+        ).frames[0]
+
+    output_path = os.path.join(tempfile.mkdtemp(), "output.mp4")
+    export_to_video(output, output_path, fps=30)
+    return output_path, seed
+
+
+examples = [
+    "A hand enters the frame, pulling a sheet of plastic wrap over three balls of dough placed on a wooden surface. The plastic wrap is stretched to cover the dough more securely. The hand adjusts the wrap, ensuring that it is tight and smooth over the dough. The scene focuses on the hand’s movements as it secures the edges of the plastic wrap. No new objects appear, and the camera remains stationary, focusing on the action of covering the dough.",
+    "A vintage train snakes through the mountains, its plume of white steam rising dramatically against the jagged peaks. The cars glint in the late afternoon sun, their deep crimson and gold accents lending a touch of elegance. The tracks carve a precarious path along the cliffside, revealing glimpses of a roaring river far below. Inside, passengers peer out the large windows, their faces lit with awe as the landscape unfolds.",
+    "A crowded rooftop bar buzzes with energy, the city skyline twinkling like a field of stars in the background. Strings of fairy lights hang above, casting a warm, golden glow over the scene. Groups of people gather around high tables, their laughter blending with the soft rhythm of live jazz. The aroma of freshly mixed cocktails and charred appetizers wafts through the air, mingling with the cool night breeze.",
+]
+
+args = init_args()
+pipe = load_model(args)
+print("load model successfully")
+with gr.Blocks() as demo:
+    gr.Markdown("# Fastvideo Mochi Video Generation Demo")
+
+    with gr.Group():
+        with gr.Row():
+            prompt = gr.Text(
+                label="Prompt",
+                show_label=False,
+                max_lines=1,
+                placeholder="Enter your prompt",
+                container=False,
+            )
+            run_button = gr.Button("Run", scale=0)
+        result = gr.Video(label="Result", show_label=False)
+
+    with gr.Accordion("Advanced options", open=False):
+        with gr.Group():
+            with gr.Row():
+                height = gr.Slider(
+                    label="Height",
+                    minimum=256,
+                    maximum=1024,
+                    step=32,
+                    value=args.height,
+                )
+                width = gr.Slider(label="Width", minimum=256, maximum=1024, step=32, value=args.width)
+
+            with gr.Row():
+                num_frames = gr.Slider(
+                    label="Number of Frames",
+                    minimum=21,
+                    maximum=163,
+                    value=args.num_frames,
+                )
+                guidance_scale = gr.Slider(
+                    label="Guidance Scale",
+                    minimum=1,
+                    maximum=12,
+                    value=args.guidance_scale,
+                )
+                num_inference_steps = gr.Slider(
+                    label="Inference Steps",
+                    minimum=4,
+                    maximum=100,
+                    value=args.num_inference_steps,
+                )
+
+            with gr.Row():
+                use_negative_prompt = gr.Checkbox(label="Use negative prompt", value=False)
+            negative_prompt = gr.Text(
+                label="Negative prompt",
+                max_lines=1,
+                placeholder="Enter a negative prompt",
+                visible=False,
+            )
+
+            seed = gr.Slider(label="Seed", minimum=0, maximum=1000000, step=1, value=args.seed)
+            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+            seed_output = gr.Number(label="Used Seed")
+
+    gr.Examples(examples=examples, inputs=prompt)
+
+    use_negative_prompt.change(
+        fn=lambda x: gr.update(visible=x),
+        inputs=use_negative_prompt,
+        outputs=negative_prompt,
+    )
+
+    run_button.click(
+        fn=generate_video,
+        inputs=[
+            prompt,
+            negative_prompt,
+            use_negative_prompt,
+            seed,
+            guidance_scale,
+            num_frames,
+            height,
+            width,
+            num_inference_steps,
+            randomize_seed,
+        ],
+        outputs=[result, seed_output],
+    )
+
+if __name__ == "__main__":
+    demo.queue(max_size=20).launch(server_name="0.0.0.0", server_port=7860)
@@ -0,0 +1,15 @@
+Fast-Hunyuan comparison with original Hunyuan, achieving an 8X diffusion speed boost with the FastVideo framework.
+
+https://github.com/user-attachments/assets/064ac1d2-11ed-4a0c-955b-4d412a96ef30
+
+Fast-Mochi comparison with original Mochi, achieving an 8X diffusion speed boost with the FastVideo framework.
+
+https://github.com/user-attachments/assets/5fbc4596-56d6-43aa-98e0-da472cf8e26c
+
+Comparison between OpenAI Sora, original Hunyuan and FastHunyuan
+
+https://github.com/user-attachments/assets/d323b712-3f68-42b2-952b-94f6a49c4836
+
+Comparison between original FastHunyuan, LLM-INT8 quantized FastHunyuan and NF4 quantized FastHunyuan
+
+https://github.com/user-attachments/assets/cf89efb5-5f68-4949-a085-f41c1ef26c94
@@ -43,7 +43,7 @@ RUN source $HOME/.local/bin/env && \
    source /opt/venv/bin/activate && \
    uv pip install --no-cache-dir --upgrade pip && \
    uv pip install --no-cache-dir .[dev] && \
-    uv pip install --no-cache-dir https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.7.16/flash_attn-2.8.3+cu128torch2.10-cp310-cp310-linux_x86_64.whl
+    uv pip install --no-cache-dir https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.5.4/flash_attn-2.8.3%2Bcu128torch2.9-cp310-cp310-linux_x86_64.whl

 COPY . .

@@ -43,7 +43,7 @@ RUN source $HOME/.local/bin/env && \
    source /opt/venv/bin/activate && \
    uv pip install --no-cache-dir --upgrade pip && \
    uv pip install --no-cache-dir .[dev] && \
-    uv pip install --no-cache-dir https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.7.16/flash_attn-2.8.3+cu128torch2.10-cp311-cp311-linux_x86_64.whl
+    uv pip install --no-cache-dir https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.5.4/flash_attn-2.8.3%2Bcu128torch2.9-cp311-cp311-linux_x86_64.whl

 COPY . .

@@ -43,7 +43,7 @@ RUN source $HOME/.local/bin/env && \
    source /opt/venv/bin/activate && \
    uv pip install --no-cache-dir --upgrade pip && \
    uv pip install --no-cache-dir .[dev] && \
-    uv pip install --no-cache-dir https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.7.16/flash_attn-2.8.3+cu128torch2.10-cp312-cp312-linux_x86_64.whl
+    uv pip install --no-cache-dir https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.5.4/flash_attn-2.8.3%2Bcu128torch2.9-cp312-cp312-linux_x86_64.whl

 COPY . .

@@ -121,15 +121,6 @@ This page contains the complete API reference for the FastVideo library.
      show_root_toc_entry: true
      heading_level: 4

-## fastvideo.registry
-
-::: fastvideo.registry
-    options:
-      show_source: true
-      show_root_heading: true
-      show_root_toc_entry: true
-      heading_level: 3
-
 ## fastvideo.pipelines

 ::: fastvideo.pipelines
@@ -324,8 +324,8 @@ Purpose:
 Action:

 - Add a new pipeline config + sampling params.
- Register them in `fastvideo/registry.py` using explicit
-  `register_configs(...)` blocks (this file is the single source of truth now).
+- Register them in `fastvideo/configs/pipelines/registry.py` and
+  `fastvideo/configs/sample/registry.py`.

 ### 5) Wire pipeline stages

@@ -17,7 +17,7 @@ You can easily use the FastVideo Pod Template on [RunPod](https://www.runpod.io)
 - Select the "FastVideo" or "fastvideo-dev" Pod Template.
 ![RunPod Pod Template Selection](../../assets/images/runpod_create.png)

- Set the Pod name to "`<name>-<FastVideo>-<date>`".
+- Set the Pod name to "<name>-<FastVideo>-<date>".

 - Finally, once the pod is deployed (will take a few minutes as the image is being pulled), you can SSH into it using "SSH exposed over TCP". You'll need to use the matching private ssh key you provided.
 ![RunPod SSH](../../assets/images/runpod_ssh.png)
@@ -51,9 +51,8 @@ runtime parameters consistent:
 - `fastvideo/configs/pipelines/`: pipeline wiring and required components.
 - `fastvideo/configs/sample/`: default sampling parameters (steps, frames,
  guidance scale, resolution, fps).
- `fastvideo/registry.py`: unified registry for pipeline config + sampling
-  defaults and model metadata resolution, defined via explicit
-  `register_configs(...)` blocks (no separate dict registries).
+- `fastvideo/configs/registry.py`: pipeline registry.
+- `fastvideo/configs/sample/registry.py`: sampling param registry.

 `FastVideoArgs` (in `fastvideo/fastvideo_args.py`) provides runtime settings and
 is passed into pipeline construction and stages.
@@ -131,41 +131,17 @@ self.attn = DistributedAttention(

 ### Registering Models

-Register implemented modules for auto‑discovery by adding `EntryClass` in each
-model module (the registry scans for it):
+Register implemented modules in the model registry:

 ```python
-# In fastvideo/models/dits/your_module.py
-class YourTransformerModel(...):
-    ...
+# In fastvideo/models/registry.py
+_TEXT_TO_VIDEO_DIT_MODELS = {
+    "YourTransformerModel": ("dits", "yourmodule", "YourTransformerClass"),
+}

-# Entry point for model registry
-EntryClass = YourTransformerModel
-```
-
-```python
-# In fastvideo/models/vaes/your_vae.py
-class YourVAEModel(...):
-    ...
-
-# Entry point for model registry
-EntryClass = YourVAEModel
-```
-
-Register pipeline config + sampling defaults in the unified registry:
-
-```python
-# In fastvideo/registry.py
-register_configs(
-    sampling_param_cls=YourSamplingParam,
-    pipeline_config_cls=YourPipelineConfig,
-    hf_model_paths=[
-        "org/your-model-id",
-    ],
-    model_detectors=[
-        lambda path: "your-model" in path.lower(),
-    ],
-)
+_VAE_MODELS = {
+    "YourVAEModel": ("vaes", "yourvae", "YourVAEClass"),
+}
 ```

 ## Step 2: Directory Structure
@@ -103,7 +103,7 @@ python setup.py install  # or pip install -e .

 **`SAGE_ATTN_THREE`**

-[SageAttention 3](https://github.com/thu-ml/SageAttention/tree/main/sageattention3_blackwell) is an advanced attention mechanism that leverages FP4 quantization and Blackwell GPU Tensor Cores for significant performance improvements.
+[SageAttention 3](https://huggingface.co/jt-zhang/SageAttention3) is an advanced attention mechanism that leverages FP4 quantization and Blackwell GPU Tensor Cores for significant performance improvements.

 #### Hardware Requirements

@@ -113,7 +113,11 @@ python setup.py install  # or pip install -e .

 Note that Sage Attention 3 requires `python>=3.13`, `torch>=2.8.0`, `CUDA >=12.8`. If you are using `uv` and using `torch==2.8.0` make sure that `sentencepiece==0.2.1` in the pyproject.toml file.

-To use Sage Attention 3 in FastVideo, follow the `README.md` in the linked repository to install the package from source.
+To use Sage Attention 3 in FastVideo, first get access to the SageAttention3 code, then move `sageattn/` and `setup.py` to the directory `fastvideo/attention/backends`, then install from using:
+
+```bash
+python setup.py install
+```

 ## Teacache

@@ -0,0 +1,172 @@
+#!/bin/bash
+#SBATCH --job-name=dmd_3333
+#SBATCH --partition=main
+#SBATCH --nodes=4
+#SBATCH --ntasks=4
+#SBATCH --ntasks-per-node=1
+#SBATCH --gres=gpu:8
+#SBATCH --cpus-per-task=128
+#SBATCH --mem=1440G
+#SBATCH --output=dmd_3333_output/dmd_3333_%j.out
+#SBATCH --error=dmd_3333_output/dmd_3333_%j.err
+#SBATCH --exclusive
+
+# Basic Info
+export NCCL_P2P_DISABLE=1
+export TORCH_NCCL_ENABLE_MONITORING=0
+export NCCL_DEBUG_SUBSYS=INIT,NET
+# different cache dir for different processes
+export TRITON_CACHE_DIR=/tmp/triton_cache_${SLURM_PROCID}
+export MASTER_PORT=29500
+export NODE_RANK=$SLURM_PROCID
+nodes=( $(scontrol show hostnames $SLURM_JOB_NODELIST) )
+export MASTER_ADDR=${nodes[0]}
+export TOKENIZERS_PARALLELISM=false
+# export WANDB_API_KEY="8d9f4b39abd68eb4e29f6fc010b7ee71a2207cde"
+export WANDB_API_KEY="50632ebd88ffd970521cec9ab4a1a2d7e85bfc45"
+# export WANDB_API_KEY='your_wandb_api_key_here'
+export WANDB_BASE_URL="https://api.wandb.ai"
+export WANDB_MODE=online
+export FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN
+
+source ~/conda/miniconda/bin/activate
+conda activate wei-fv-distill
+export HOME="/mnt/weka/home/hao.zhang/wei"
+
+# Configs
+NUM_GPUS=8
+NUM_TOTAL_GPUS=32
+
+# Model paths for Self-Forcing DMD distillation with Hunyuan1.5:
+GENERATOR_MODEL_PATH="weizhou03/SFHunyuanVideo-1.5-Diffusers-480p_t2v"  # Updated to Wan2.2
+REAL_SCORE_MODEL_PATH="hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v" # Teacher model
+FAKE_SCORE_MODEL_PATH="hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v"
+# REAL_SCORE_MODEL_PATH="hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_i2v_step_distilled"
+# FAKE_SCORE_MODEL_PATH="hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_i2v_step_distilled" # Critic model
+
+# DATA_DIR="data/test-text-preprocessing/Node_0_GPU_1_File_1/combined_parquet_dataset/"
+# DATA_DIR="/mnt/weka/home/hao.zhang/matthew/FastVideo/data/test-text-preprocessing"
+DATA_DIR="/mnt/weka/home/hao.zhang/wl/sharefs/wl/release/vidprom_16k_text_embed"
+DATA_DIR_2="/mnt/weka/home/hao.zhang/wl/sharefs/wl/release/fv-ode-preprocessing-16k-hy15-121"
+# DATA_DIR=data/crush-smol_processed_t2v/combined_parquet_dataset
+# DATA_DIR="/mnt/sharefs/users/hao.zhang/Vchitect-2M/Wan-Syn-upload/latents_i2v/train/"
+# VALIDATION_DATASET_FILE="data/crush-smol-single_processed_t2v/validation.json"
+VALIDATION_DATASET_FILE="/mnt/weka/home/hao.zhang/wl/FastVideo/examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free/validation_64.json"
+
+# export CUDA_VISIBLE_DEVICES=4,5
+# IP=[MASTER NODE IP]
+
+training_args=(
+  --tracker_project_name SFhy1.5_t2v_distill_self_forcing_dmd  # Updated for Wan2.2
+  --output_dir "/mnt/weka/home/hao.zhang/wei/SFhy1.5_distill_3333_1e-5_1e-5_cfg6_corrected_scheduler"
+  --wandb_run_name "self_forcing_3333_context_forcing"
+  --max_train_steps 4000
+  --train_batch_size 1
+  --train_sp_batch_size 1
+  --gradient_accumulation_steps 1
+  --num_latent_t 21
+  --num_height 480
+  --num_width 848
+  --enable_gradient_checkpointing_type "full"
+  --simulate_generator_forward
+  --log_visualization
+  --num_frames 81
+  --num_frame_per_block 3  # Frame generation block size for self-forcing
+  # --resume-from-checkpoint "/mnt/weka/home/hao.zhang/wei/SFhy1.5_distill_1333/checkpoint-300"
+  --init_weights_from_safetensors /mnt/weka/home/hao.zhang/wei/hy15_worldplay_tf_init_3333/checkpoint-2800/transformer/diffusion_pytorch_model.safetensors
+  # --init_weights_from_safetensors /mnt/weka/home/hao.zhang/wei/hy15_worldplay_df_init_1333/checkpoint-11600/transformer/diffusion_pytorch_model.safetensors
+  # --init_weights_from_safetensors /mnt/weka/home/hao.zhang/wei/hy15_distilled_ode_init_3333/checkpoint-2400/transformer/diffusion_pytorch_model.safetensors
+)
+
+parallel_args=(
+  --num_gpus $NUM_TOTAL_GPUS # 64
+  --sp_size 1
+  --tp_size 1
+  --hsdp_replicate_dim 1 # 64
+  --hsdp_shard_dim $NUM_TOTAL_GPUS
+)
+
+model_args=(
+  --model_path $GENERATOR_MODEL_PATH
+  --pretrained_model_name_or_path $GENERATOR_MODEL_PATH
+  --real_score_model_path $REAL_SCORE_MODEL_PATH
+  --fake_score_model_path $FAKE_SCORE_MODEL_PATH
+)
+
+dataset_args=(
+  --data_path "$DATA_DIR"
+  # --data_path_2 "$DATA_DIR_2"
+  --dataloader_num_workers 4
+)
+
+validation_args=(
+  --log_validation
+  --validation_dataset_file "$VALIDATION_DATASET_FILE"
+  --validation_steps 50
+  --validation_sampling_steps "8"
+  --validation_guidance_scale "1.0" # not used for dmd inference
+  --text-encoder-cpu-offload
+  # --vae_cpu_offload True
+)
+
+optimizer_args=(
+  --optimizer-type "muon"
+  --learning_rate 1e-5
+  --mixed_precision "bf16"
+  --training_state_checkpointing_steps 100
+  --weight_only_checkpointing_steps 500
+  --weight_decay 0.01
+  --betas '0.0,0.999'
+  --max_grad_norm 1.0
+)
+
+miscellaneous_args=(
+  --inference_mode False
+  --checkpoints_total_limit 3
+  --training_cfg_rate 0.0
+  --dit_precision "fp32"
+  --flow_shift 5
+  --seed 1000
+  --use_ema True
+  --ema_decay 0.99
+  --ema_start_step 200
+)
+
+dmd_args=(
+  --warp_denoising_step
+  # --dmd_denoising_steps '1000,875,750,625,500,375,250,125'
+  # --dmd_denoising_steps '1000,760,520,280'
+  # --dmd_denoising_steps '1000,750,500,250'
+  --min_timestep_ratio 0.02
+  --max_timestep_ratio 0.98
+  --dfake_gen_update_ratio 5
+  --real_score_guidance_scale 3.5
+  --fake_score_learning_rate 5e-6
+  --fake_score_betas '0.0,0.999'
+)
+
+self_forcing_args=(
+  --independent_first_frame False  # Whether to treat first frame independently
+  --same_step_across_blocks True  # Whether to use same denoising step across all blocks
+  --last_step_only False  # Whether to only use the last denoising step
+  --context_noise 0  # Amount of noise to add during context caching (0 = no noise)
+  # --use-ode-init True
+  # --use-context-forcing True
+)
+
+srun torchrun \
+--nnodes $SLURM_JOB_NUM_NODES \
+--nproc_per_node $NUM_GPUS \
+--node_rank $SLURM_PROCID \
+--rdzv_backend=c10d \
+--rdzv_endpoint="$MASTER_ADDR:$MASTER_PORT" \
+    fastvideo/training/hy15_self_forcing_distillation_pipeline.py \
+    "${parallel_args[@]}" \
+    "${model_args[@]}" \
+    "${dataset_args[@]}" \
+    "${training_args[@]}" \
+    "${optimizer_args[@]}" \
+    "${validation_args[@]}" \
+    "${miscellaneous_args[@]}" \
+    "${dmd_args[@]}" \
+    "${self_forcing_args[@]}"
@@ -0,0 +1,172 @@
+#!/bin/bash
+#SBATCH --job-name=distill_dmd_t2v_1.3b
+#SBATCH --nodes=8
+#SBATCH --ntasks=8
+#SBATCH --ntasks-per-node=1
+#SBATCH --gres=gpu:8
+#SBATCH --cpus-per-task=128
+#SBATCH --mem=1440G
+#SBATCH --output=distill_dmd_t2v_1.3b_output/distill_dmd_t2v_1.3b_%j.out
+#SBATCH --error=distill_dmd_t2v_1.3b_output/distill_dmd_t2v_1.3b_%j.err
+#SBATCH --exclusive
+
+# Basic Info
+export NCCL_P2P_DISABLE=1
+export TORCH_NCCL_ENABLE_MONITORING=0
+export NCCL_DEBUG_SUBSYS=INIT,NET
+# different cache dir for different processes
+# export TRITON_CACHE_DIR=/tmp/triton_cache_${SLURM_PROCID}
+export MASTER_PORT=29500
+export NODE_RANK=$SLURM_PROCID
+nodes=( $(scontrol show hostnames $SLURM_JOB_NODELIST) )
+export MASTER_ADDR=${nodes[0]}
+export TOKENIZERS_PARALLELISM=false
+# export WANDB_API_KEY="8d9f4b39abd68eb4e29f6fc010b7ee71a2207cde"
+export WANDB_API_KEY="50632ebd88ffd970521cec9ab4a1a2d7e85bfc45"
+# export WANDB_API_KEY='your_wandb_api_key_here'
+export WANDB_BASE_URL="https://api.wandb.ai"
+export WANDB_MODE=online
+export FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN
+
+export HF_HOME=/mnt/data/huggingface
+export HF_DATASETS_CACHE="/tmp/hf_datasets_${SLURM_JOB_ID}_${SLURM_PROCID}"
+mkdir -p "${HF_DATASETS_CACHE}"
+export CUDA_HOME=/usr/local/cuda
+export PATH="$CUDA_HOME/bin:$PATH"
+export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/home/shared-bin/local/usr/lib/aarch64-linux-gnu:$CUDA_HOME/lib64"
+export CUDACXX="$CUDA_HOME/bin/nvcc"
+
+source ~/miniconda3/bin/activate
+conda activate fastvideo
+
+# Configs
+NUM_GPUS=8
+NUM_TOTAL_GPUS=64
+
+# Model paths for Self-Forcing DMD distillation:
+GENERATOR_MODEL_PATH="wlsaidhi/SFWan2.1-T2V-1.3B-Diffusers"
+REAL_SCORE_MODEL_PATH="Wan-AI/Wan2.1-T2V-14B-Diffusers"  # Teacher model
+FAKE_SCORE_MODEL_PATH="Wan-AI/Wan2.1-T2V-1.3B-Diffusers"  # Critic model
+
+DATA_DIR="/mnt/data/vidprom_filtered_extended_umt5_text_embed"
+# DATA_DIR="/mnt/data/mixkit_wan_1.3b_processed_t2v_distributed"
+DATA_DIR_2="/mnt/data/mixkit_wan_1.3b_processed_t2v_distributed"
+VALIDATION_DATASET_FILE="/mnt/home/zhouw.jerry2017/FastVideo/examples/distill/SFWan2.1-T2V/validation_64.json"
+# export CUDA_VISIBLE_DEVICES=4,5
+# IP=[MASTER NODE IP]
+
+training_args=(
+  --tracker_project_name SFwan_t2v_distill_self_forcing_dmd
+  --wandb_run_name causal_forcing_vidprom_extended
+  --output_dir /mnt/data/wan_cf_df_distill_1.3b
+  --max_train_steps 4000
+  --train_batch_size 1
+  --train_sp_batch_size 1
+  --gradient_accumulation_steps 1
+  --num_latent_t 21
+  --num_height 480
+  --num_width 832
+  --enable_gradient_checkpointing_type "full"
+  --log_visualization
+  --simulate_generator_forward
+  --num_frames 81
+  --num_frame_per_block 3  # Frame generation block size for self-forcing
+  --enable_gradient_masking
+  --gradient_mask_last_n_frames 21
+)
+
+parallel_args=(
+  --num_gpus $NUM_TOTAL_GPUS # 64
+  --sp_size 1
+  --tp_size 1
+  --hsdp_replicate_dim 1 # 64
+  --hsdp_shard_dim $NUM_TOTAL_GPUS
+)
+
+model_args=(
+  --model_path $GENERATOR_MODEL_PATH  # TODO: check if you can remove this in this script
+  --pretrained_model_name_or_path $GENERATOR_MODEL_PATH
+  --real_score_model_path $REAL_SCORE_MODEL_PATH
+  --fake_score_model_path $FAKE_SCORE_MODEL_PATH
+)
+
+dataset_args=(
+  --data_path "$DATA_DIR"
+  # --data_path_2 "$DATA_DIR_2"
+  --dataloader_num_workers 4
+)
+
+validation_args=(
+  --log_validation
+  --validation_dataset_file "$VALIDATION_DATASET_FILE"
+  --validation_steps 100
+  --validation_sampling_steps "4"
+  --validation_guidance_scale "1.0"
+)
+
+optimizer_args=(
+  --optimizer-type "adamw"
+  --learning_rate 1e-5
+  --mixed_precision "bf16"
+  --training_state_checkpointing_steps 500
+  --weight_only_checkpointing_steps 500
+  --weight_decay 0.01
+  --betas '0.0,0.999'
+  --max_grad_norm 1.0
+)
+
+miscellaneous_args=(
+  --inference_mode False
+  --checkpoints_total_limit 3
+  --training_cfg_rate 0.0
+  --dit_precision "fp32"
+  --flow_shift 5
+  --seed 1000
+  --use_ema True
+  --ema_decay 0.99
+  --ema_start_step 200
+#   --init_weights_from_safetensors "/mnt/data/wan_tf_ode_init_3333/checkpoint-1600/transformer/diffusion_pytorch_model.safetensors"
+#   --init_weights_from_safetensors "/mnt/data/wan_tf_init_3333/checkpoint-800/transformer/diffusion_pytorch_model.safetensors"
+  # --init_weights_from_safetensors "/mnt/data/Causal-Forcing-wan1.3b/causal_ode.safetensors"
+  # --init_weights_from_safetensors "/mnt/data/wan_ar_diffusion_3333_mixkit_shift5/checkpoint-7200/transformer/diffusion_pytorch_model.safetensors"
+  --init_weights_from_safetensors "/mnt/data/wan_ar_diffusion_3333_mixkit_shift5/checkpoint-4800/transformer/diffusion_pytorch_model.safetensors"
+  # --init_weights_from_safetensors "/mnt/data/wan_tf_ode_init_3333_mixkit_shift5/checkpoint-2000/transformer/diffusion_pytorch_model.safetensors"
+  # --init_weights_from_safetensors "/mnt/data/wan_ar_df_diffusion_3333_mixkit_shift5/checkpoint-4800/transformer/diffusion_pytorch_model.safetensors"
+)
+
+dmd_args=(
+  --dmd_denoising_steps '1000,750,500,250'
+  --min_timestep_ratio 0.02
+  --max_timestep_ratio 0.98
+  --dfake_gen_update_ratio 5
+  --real_score_guidance_scale 3.0
+  --fake_score_learning_rate 2e-6
+  --fake_score_betas '0.0,0.999'
+  --warp_denoising_step
+  # --use-decoupled-dmd True
+  # --use-gan-loss True
+)
+
+self_forcing_args=(
+  --independent_first_frame False  # Whether to treat first frame independently
+  --same_step_across_blocks True  # Whether to use same denoising step across all blocks
+  --last_step_only False  # Whether to only use the last denoising step
+  --context_noise 0  # Amount of noise to add during context caching (0 = no noise)
+)
+
+srun torchrun \
+--nnodes $SLURM_JOB_NUM_NODES \
+--nproc_per_node $NUM_GPUS \
+--node_rank $SLURM_PROCID \
+--rdzv_backend=c10d \
+--rdzv_endpoint="$MASTER_ADDR:$MASTER_PORT" \
+    fastvideo/training/wan_self_forcing_distillation_pipeline.py \
+    "${parallel_args[@]}" \
+    "${model_args[@]}" \
+    "${dataset_args[@]}" \
+    "${training_args[@]}" \
+    "${optimizer_args[@]}" \
+    "${validation_args[@]}" \
+    "${miscellaneous_args[@]}" \
+    "${dmd_args[@]}" \
+    "${self_forcing_args[@]}"
@@ -15,7 +15,7 @@ def main():
        use_fsdp_inference=False, # set to True if GPU is out of memory
        dit_cpu_offload=False,
        vae_cpu_offload=False,
-        text_encoder_cpu_offload=True,
+        text_encoder_cpu_offload=False,
        pin_cpu_memory=True, # set to false if low CPU RAM or hit obscure "CUDA error: Invalid argument"
        # image_encoder_cpu_offload=False,
    )
@@ -1,50 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-from fastvideo import VideoGenerator
-from fastvideo.configs.sample import SamplingParam
-
-
-def main():
-    # Point this to your local diffusers model dir (or replace with a HF model ID).
-    model_path = "KyleShao/Cosmos-Predict2.5-2B-Diffusers"
-
-    generator = VideoGenerator.from_pretrained(
-        model_path,
-        num_gpus=1,
-        use_fsdp_inference=False,  # set True if GPU is out of memory
-        dit_cpu_offload=False,
-        vae_cpu_offload=False,
-        text_encoder_cpu_offload=True,
-        pin_cpu_memory=True,
-    )
-
-    sampling_param = SamplingParam.from_pretrained(model_path)
-
-    # image2world example from official repo
-    image_path = "assets/images/bus_terminal.jpg"
-
-    prompt = (
-        "A nighttime city bus terminal gradually shifts from stillness to subtle movement. "
-        "At first, multiple double-decker buses are parked under the glow of overhead lights, "
-        "with a central bus labeled '87D' facing forward and stationary. "
-        "As the video progresses, the bus in the middle moves ahead slowly, its headlights brightening the surrounding area "
-        "and casting reflections onto adjacent vehicles. "
-        "The motion creates space in the lineup, signaling activity within the otherwise quiet station. "
-        "It then comes to a smooth stop, resuming its position in line. "
-        "Overhead signage in Chinese characters remains illuminated, enhancing the vibrant, urban night scene."
-    )
-
-    generator.generate_video(
-        prompt,
-        sampling_param=sampling_param,
-        image_path=str(image_path),
-        num_cond_frames=1,
-        output_path="outputs_video/cosmos2_5_i2w.mp4",
-        save_video=True,
-    )
-
-    generator.shutdown()
-
-
-if __name__ == "__main__":
-    main()
-
@@ -1,5 +1,4 @@
 from fastvideo import VideoGenerator
-from fastvideo.configs.sample import SamplingParam


 def main():
@@ -16,27 +15,19 @@ def main():
        pin_cpu_memory=True,
    )

-    # Load default sampling parameters (negative_prompt, resolution, steps, etc.)
-    sampling_param = SamplingParam.from_pretrained(model_path)
-
    prompt = (
-        "A high-definition video captures the precision of robotic welding in an industrial setting. "
-        "The first frame showcases a robotic arm, equipped with a welding torch, positioned over a large metal structure. "
-        "The welding process is in full swing, with bright sparks and intense light illuminating the scene, "
-        "creating a vivid display of blue and white hues. "
-        "A significant amount of smoke billows around the welding area, partially obscuring the view but emphasizing the heat and activity. "
-        "The background reveals parts of the workshop environment, including a ventilation system and various pieces of machinery, "
-        "indicating a busy and functional industrial workspace. "
-        "As the video progresses, the robotic arm maintains its steady position, continuing the welding process and moving to its left. "
-        "The welding torch consistently emits sparks and light, and the smoke continues to rise, diffusing slightly as it moves upward. "
-        "The metal surface beneath the torch shows ongoing signs of heating and melting. "
-        "The scene retains its industrial ambiance, with the welding sparks and smoke dominating the visual field, "
-        "underscoring the ongoing nature of the welding operation."
+       "A high-definition video captures the precision of robotic welding in an industrial setting. The first frame showcases a robotic arm, equipped with a welding torch, positioned over a large metal structure. The welding process is in full swing, with bright sparks and intense light illuminating the scene, creating a vivid display of blue and white hues. A significant amount of smoke billows around the welding area, partially obscuring the view but emphasizing the heat and activity. The background reveals parts of the workshop environment, including a ventilation system and various pieces of machinery, indicating a busy and functional industrial workspace. As the video progresses, the robotic arm maintains its steady position, continuing the welding process and moving to its left. The welding torch consistently emits sparks and light, and the smoke continues to rise, diffusing slightly as it moves upward. The metal surface beneath the torch shows ongoing signs of heating and melting. The scene retains its industrial ambiance, with the welding sparks and smoke dominating the visual field, underscoring the ongoing nature of the welding operation."
    )

-    generator.generate_video(
+    video = generator.generate_video(
        prompt,
-        sampling_param=sampling_param,
+        negative_prompt="",
+        height=704,
+        width=1280,
+        num_frames=77,
+        num_inference_steps=35,
+        guidance_scale=7.0,
+        fps=24,
        output_path="outputs_video/cosmos2_5_t2w.mp4",
        save_video=True,
    )
@@ -1,53 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-from fastvideo import VideoGenerator
-from fastvideo.configs.sample import SamplingParam
-
-
-def main():
-    # Point this to your local diffusers model dir (or replace with a HF model ID).
-    model_path = "KyleShao/Cosmos-Predict2.5-2B-Diffusers"
-
-    generator = VideoGenerator.from_pretrained(
-        model_path,
-        num_gpus=1,
-        use_fsdp_inference=False,  # set True if GPU is out of memory
-        dit_cpu_offload=False,
-        vae_cpu_offload=False,
-        text_encoder_cpu_offload=True,
-        pin_cpu_memory=True,
-    )
-
-    sampling_param = SamplingParam.from_pretrained(model_path)
-
-    # video2world example from official repo
-    video_path = "assets/videos/robot_pouring.mp4"
-
-    prompt = (
-        "A robotic arm, primarily white with black joints and cables, is shown in a clean, modern indoor setting with a white tabletop. "
-        "The arm, equipped with a gripper holding a small, light green pitcher, is positioned above a clear glass containing a reddish-brown liquid and a spoon. "
-        "The robotic arm is in the process of pouring a transparent liquid into the glass. "
-        "To the left of the pitcher, there is an opened jar with a similar reddish-brown substance visible through its transparent body. "
-        "In the background, a vase with white flowers and a brown couch are partially visible, adding to the contemporary ambiance. "
-        "The lighting is bright, casting soft shadows on the table. "
-        "The robotic arm's movements are smooth and controlled, demonstrating precision in its task. "
-        "As the video progresses, the robotic arm completes the pour, leaving the glass half-filled with the reddish-brown liquid. "
-        "The jar remains untouched throughout the sequence, and the spoon inside the glass remains stationary. "
-        "The other robotic arm on the right side also stays stationary throughout the video. "
-        "The final frame captures the robotic arm with the pitcher finishing the pour, with the glass now filled to a higher level, while the pitcher is slightly tilted but still held securely by the gripper."
-    )
-
-    generator.generate_video(
-        prompt,
-        sampling_param=sampling_param,
-        video_path=str(video_path),
-        num_cond_frames=1,
-        output_path="outputs_video/cosmos2_5_v2w.mp4",
-        save_video=True,
-    )
-
-    generator.shutdown()
-
-
-if __name__ == "__main__":
-    main()
-
@@ -1,119 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-"""
-Basic inference script for HunyuanGameCraft video generation.
-
-HunyuanGameCraft generates game-like videos with camera/action control.
-It takes an optional image input and generates video with camera motion
-based on simple action commands (forward, left, right, backward, rotations).
-
-Available actions:
-    - forward (w): Move camera forward
-    - backward (s): Move camera backward  
-    - left (a): Move camera left (strafe)
-    - right (d): Move camera right (strafe)
-    - left_rot: Rotate camera left (pan)
-    - right_rot: Rotate camera right (pan)
-    - up_rot: Rotate camera up (tilt)
-    - down_rot: Rotate camera down (tilt)
-
-T2V vs I2V:
-    - Default: I2V (uses a default reference image). Set GAMECRAFT_I2V_IMAGE to a
-      URL or path to use a different image.
-    - T2V only (no reference image): run with GAMECRAFT_I2V_IMAGE= (empty).
-"""
-import os
-
-import torch
-
-from fastvideo import VideoGenerator
-from fastvideo.models.camera import create_camera_trajectory
-
-# Model configuration (use GAMECRAFT_MODEL_PATH for local weights)
-MODEL_PATH = os.environ.get("GAMECRAFT_MODEL_PATH", "FastVideo/HunyuanGameCraft-Diffusers")
-
-# Default prompts for demo
-DEFAULT_PROMPTS = {
-    "village": "A charming medieval village with cobblestone streets, thatched-roof houses, and vibrant flower gardens under a bright blue sky.",
-    "temple": "A majestic ancient temple stands under a clear blue sky, its grandeur highlighted by towering Doric columns and intricate architectural details.",
-    "forest": "A lush green forest with tall trees, dappled sunlight filtering through the leaves, and a winding dirt path.",
-    "beach": "A tropical beach with crystal clear turquoise water, white sand, and palm trees swaying in the breeze.",
-}
-
-# I2V: default reference image (URL). Can override with a local path.
-DEFAULT_I2V_IMAGE_URL = (
-    "https://huggingface.co/datasets/huggingface/documentation-images/"
-    "resolve/main/diffusers/astronaut.jpg"
-)
-DEFAULT_I2V_PROMPT = (
-    "An astronaut hatching from an egg, on the surface of the moon, "
-    "the darkness and depth of space realised in the background."
-)
-
-OUTPUT_PATH = "video_samples_gamecraft"
-
-
-def main():
-    # Initialize generator
-    # FastVideo will automatically download weights from HuggingFace
-    generator = VideoGenerator.from_pretrained(
-        MODEL_PATH,
-        num_gpus=1,
-        use_fsdp_inference=True,
-        dit_cpu_offload=True,
-        vae_cpu_offload=True,
-        text_encoder_cpu_offload=True,
-        pin_cpu_memory=True,
-    )
-
-    # Video parameters
-    height = 704
-    width = 1280
-    num_frames = 33
-    action = "forward"
-    action_speed = 0.2
-
-    # Create camera trajectory (Plücker coordinates)
-    camera_states = create_camera_trajectory(
-        action=action,
-        height=height,
-        width=width,
-        num_frames=num_frames,
-        action_speed=action_speed,
-        dtype=torch.bfloat16,
-    )
-    print(f"Camera states shape: {camera_states.shape}")
-
-    # I2V vs T2V: unset GAMECRAFT_I2V_IMAGE -> I2V (default image). Set to "" -> T2V.
-    env_image = os.environ.get("GAMECRAFT_I2V_IMAGE")
-    if env_image is None:
-        image_path = DEFAULT_I2V_IMAGE_URL  # default: I2V
-    elif env_image.strip() == "":
-        image_path = None  # T2V
-    else:
-        image_path = env_image.strip()  # I2V with given URL/path
-
-    is_i2v = image_path is not None
-    prompt = DEFAULT_I2V_PROMPT if is_i2v else DEFAULT_PROMPTS["temple"]
-    print(f"Mode: {'I2V' if is_i2v else 'T2V'}, prompt: {prompt[:60]}...")
-
-    gen_kw = dict(
-        prompt=prompt,
-        negative_prompt="",
-        camera_states=camera_states,
-        height=height,
-        width=width,
-        num_frames=num_frames,
-        num_inference_steps=50,
-        guidance_scale=6.0,
-        seed=42,
-        fps=24,
-        output_path=OUTPUT_PATH,
-        save_video=True,
-    )
-    if is_i2v:
-        gen_kw["image_path"] = image_path
-    generator.generate_video(**gen_kw)
-
-
-if __name__ == "__main__":
-    main()
@@ -39,4 +39,4 @@ def main():


 if __name__ == "__main__":
-    main()
+    main()
@@ -1,43 +0,0 @@
-from fastvideo import VideoGenerator
-import json
-# from fastvideo.configs.sample import SamplingParam
-
-OUTPUT_PATH = "video_samples_hy15_1080p"
-def main():
-    # FastVideo will automatically use the optimal default arguments for the
-    # model.
-    # If a local path is provided, FastVideo will make a best effort
-    # attempt to identify the optimal arguments.
-    generator = VideoGenerator.from_pretrained(
-        "weizhou03/HunyuanVideo-1.5-Diffusers-1080p-2SR", # 480p -> 720p -> 1080p
-        # or "weizhou03/HunyuanVideo-1.5-Diffusers-1080p" # 720p -> 1080p
-        # FastVideo will automatically handle distributed setup
-        num_gpus=1,
-        use_fsdp_inference=False, # set to True if GPU is out of memory
-        dit_cpu_offload=True,
-        vae_cpu_offload=True,
-        text_encoder_cpu_offload=True,
-        pin_cpu_memory=True, # set to false if low CPU RAM or hit obscure "CUDA error: Invalid argument"
-        # image_encoder_cpu_offload=False,
-    )
-
-    prompt = (
-        "A curious raccoon peers through a vibrant field of yellow sunflowers, its eyes "
-        "wide with interest. The playful yet serene atmosphere is complemented by soft "
-        "natural light filtering through the petals. Mid-shot, warm and cheerful tones."
-    )
-
-    video = generator.generate_video(prompt, output_path=OUTPUT_PATH, save_video=True, negative_prompt="")
-
-    prompt2 = (
-        "A majestic lion strides across the golden savanna, its powerful frame "
-        "glistening under the warm afternoon sun. The tall grass ripples gently in "
-        "the breeze, enhancing the lion's commanding presence. The tone is vibrant, "
-        "embodying the raw energy of the wild. Low angle, steady tracking shot, "
-        "cinematic.")
-
-    video2 = generator.generate_video(prompt2, output_path=OUTPUT_PATH, save_video=True, negative_prompt="")
-
-
-if __name__ == "__main__":
-    main()
@@ -1,11 +1,165 @@
+# SPDX-License-Identifier: Apache-2.0
+"""
+Basic example for HYWorld (HY-WorldPlay) video generation using FastVideo.
+
+This example replicates the same functionality as HY-WorldPlay/run.sh,
+demonstrating image-to-video generation with camera trajectory control.
+"""
+
+import time
+import math
+import numpy as np
+import imageio
+import torchvision
+from einops import rearrange
+
 from fastvideo import VideoGenerator
+from fastvideo.pipelines import ForwardBatch
+from fastvideo.utils import shallow_asdict, align_to
+from fastvideo.logger import init_logger
+
+from fastvideo.models.dits.hyworld.pose import pose_to_input, compute_latent_num
 from fastvideo.models.dits.hyworld.resolution_utils import get_resolution_from_image

+logger = init_logger(__name__)
+
+class HYWorldVideoGenerator(VideoGenerator):
+    """Extended VideoGenerator that adds HYWorld-specific parameters to batch.extra."""
+
+    def _generate_single_video(self, prompt: str, sampling_param=None, **kwargs):
+        """Override to add viewmats, Ks, and action to batch.extra."""
+        fastvideo_args = self.fastvideo_args
+        pipeline_config = fastvideo_args.pipeline_config
+
+        if sampling_param is None:
+            from fastvideo.configs.sample import SamplingParam
+            sampling_param = SamplingParam.from_pretrained(fastvideo_args.model_path)
+
+        # Update sampling param with kwargs
+        if kwargs:
+            for key, value in kwargs.items():
+                if hasattr(sampling_param, key):
+                    setattr(sampling_param, key, value)
+
+        # Get pose string from sampling_param or kwargs
+        pose = kwargs.get('pose', getattr(sampling_param, 'POSE', 'w-31'))
+        num_frames = kwargs.get('num_frames', getattr(sampling_param, 'num_frames', 125))
+
+        # Calculate number of latents
+        latent_num = compute_latent_num(num_frames)
+
+        # Convert pose to viewmats, Ks, and action
+        viewmats, Ks, action = pose_to_input(pose, latent_num)
+
+        # Convert to tensors and add batch dimension
+        viewmats = viewmats.unsqueeze(0)  # (1, T, 4, 4)
+        Ks = Ks.unsqueeze(0)  # (1, T, 3, 3)
+        action = action.unsqueeze(0)  # (1, T)
+
+        # Validate inputs
+        prompt = prompt.strip()
+        sampling_param = sampling_param.__class__(**shallow_asdict(sampling_param))
+        output_path = kwargs.get("output_path", sampling_param.output_path)
+        sampling_param.prompt = prompt
+
+        if sampling_param.negative_prompt is not None:
+            sampling_param.negative_prompt = sampling_param.negative_prompt.strip()
+
+        # Validate dimensions
+        if (sampling_param.height <= 0 or sampling_param.width <= 0 or
+            sampling_param.num_frames <= 0):
+            raise ValueError(
+                f"Height, width, and num_frames must be positive integers")
+
+        temporal_scale_factor = pipeline_config.vae_config.arch_config.temporal_compression_ratio
+        num_frames = sampling_param.num_frames
+        num_gpus = fastvideo_args.num_gpus
+        use_temporal_scaling_frames = pipeline_config.vae_config.use_temporal_scaling_frames
+
+        # Adjust number of frames based on number of GPUs
+        if use_temporal_scaling_frames:
+            orig_latent_num_frames = (num_frames - 1) // temporal_scale_factor + 1
+        else:
+            orig_latent_num_frames = sampling_param.num_frames // 17 * 3
+
+        if orig_latent_num_frames % fastvideo_args.num_gpus != 0:
+            if use_temporal_scaling_frames:
+                new_num_frames = (orig_latent_num_frames - 1) * temporal_scale_factor + 1
+            else:
+                divisor = math.lcm(3, num_gpus)
+                orig_latent_num_frames = (
+                    (orig_latent_num_frames + divisor - 1) // divisor) * divisor
+                new_num_frames = orig_latent_num_frames // 3 * 17
+
+            logger.info(
+                "Adjusting number of frames from %s to %s based on number of GPUs (%s)",
+                sampling_param.num_frames, new_num_frames, fastvideo_args.num_gpus)
+            sampling_param.num_frames = new_num_frames
+
+        # Calculate sizes
+        target_height = align_to(sampling_param.height, 16)
+        target_width = align_to(sampling_param.width, 16)
+
+        # Calculate latent sizes
+        latents_size = [(sampling_param.num_frames - 1) // 4 + 1,
+                        sampling_param.height // 8, sampling_param.width // 8]
+        n_tokens = latents_size[0] * latents_size[1] * latents_size[2]
+
+        # Prepare batch
+        batch = ForwardBatch(
+            **shallow_asdict(sampling_param),
+            eta=0.0,
+            n_tokens=n_tokens,
+            VSA_sparsity=fastvideo_args.VSA_sparsity,
+        )
+
+        # Add HYWorld-specific parameters to batch.extra
+        batch.extra['viewmats'] = viewmats
+        batch.extra['Ks'] = Ks
+        batch.extra['action'] = action
+        batch.extra['chunk_latent_frames'] = 16  # For bidirectional model
+
+        # Run inference
+        start_time = time.perf_counter()
+        output_batch = self.executor.execute_forward(batch, fastvideo_args)
+        samples = output_batch.output
+        logging_info = output_batch.logging_info
+
+        gen_time = time.perf_counter() - start_time
+        logger.info("Generated successfully in %.2f seconds", gen_time)
+
+        # Process outputs
+        videos = rearrange(samples, "b c t h w -> t b c h w")
+        frames = []
+        for x in videos:
+            x = torchvision.utils.make_grid(x, nrow=6)
+            x = x.transpose(0, 1).transpose(1, 2).squeeze(-1)
+            frames.append((x * 255).numpy().astype(np.uint8))
+
+        # Save video if requested
+        if batch.save_video:
+            imageio.mimsave(output_path, frames, fps=batch.fps, format="mp4")
+            logger.info("Saved video to %s", output_path)
+
+        if batch.return_frames:
+            return frames
+        else:
+            return {
+                "samples": samples,
+                "frames": frames,
+                "prompts": prompt,
+                "size": (target_height, target_width, batch.num_frames),
+                "generation_time": gen_time,
+                "logging_info": logging_info,
+                "trajectory": output_batch.trajectory_latents,
+                "trajectory_timesteps": output_batch.trajectory_timesteps,
+                "trajectory_decoded": output_batch.trajectory_decoded,
+            }
+
 # Default prompt from HY-WorldPlay run.sh
 DEFAULT_PROMPT = 'A paved pathway leads towards a stone arch bridge spanning a calm body of water.  Lush green trees and foliage line the path and the far bank of the water. A traditional-style pavilion with a tiered, reddish-brown roof sits on the far shore. The water reflects the surrounding greenery and the sky.  The scene is bathed in soft, natural light, creating a tranquil and serene atmosphere. The pathway is composed of large, rectangular stones, and the bridge is constructed of light gray stone.  The overall composition emphasizes the peaceful and harmonious nature of the landscape.'
 DEFAULT_IMAGE = 'https://raw.githubusercontent.com/Tencent-Hunyuan/HY-WorldPlay/main/assets/img/test.png'

-OUTPUT_PATH = "video_samples_hyworld"
 def main():
    import argparse

@@ -15,7 +169,7 @@ def main():
    parser.add_argument("--prompt", type=str, default=DEFAULT_PROMPT, help="Text prompt for video generation")
    parser.add_argument("--image", type=str, default=DEFAULT_IMAGE, help="Path or URL to input image")
    parser.add_argument("--pose", type=str, default='w-31', help="Pose string (e.g., 'a-31', 'w-31', 's-31', 'd-31')")
-    parser.add_argument("--output_path", type=str, default=OUTPUT_PATH, help="Output video path")
+    parser.add_argument("--output_path", type=str, default='video_samples_hyworld', help="Output video path")
    parser.add_argument("--num-frames", type=int, default=125, help="Number of frames")
    parser.add_argument("--seed", type=int, default=1, help="Random seed")
    parser.add_argument("--resolution", type=str, default="480p", help="Only support 480p for now")
@@ -31,7 +185,8 @@ def main():

    # Initialize generator
    print("\nInitializing VideoGenerator for HYWorld...")
-    generator = VideoGenerator.from_pretrained(
+    
+    generator = HYWorldVideoGenerator.from_pretrained(
        "FastVideo/HY-WorldPlay-Bidirectional-Diffusers",
        num_gpus=1,
        use_fsdp_inference=True,
@@ -43,12 +198,11 @@ def main():
    )

    # Generate video
-    # The pose string is automatically converted to camera matrices by the pipeline
    print("\nGenerating video...")
-    generator.generate_video(
+    start_time = time.time()
+    video = generator.generate_video(
        prompt=args.prompt,
        image_path=args.image,
-        pose=args.pose,  # Camera trajectory control
        output_path=args.output_path,
        save_video=True,
        negative_prompt="",
@@ -57,9 +211,13 @@ def main():
        height=HEIGHT,
        width=WIDTH,
        seed=args.seed,
+        pose=args.pose,
    )
+    elapsed = time.time() - start_time

-    print(f"\nVideo saved to: {args.output_path}")
+    print(f"\nVideo generated successfully!")
+    print(f"Saved to: {args.output_path}")
+    print(f"Time: {elapsed:.2f}s")


 if __name__ == "__main__":
@@ -1,49 +0,0 @@
-from fastvideo import VideoGenerator
-from fastvideo.models.dits.lingbotworld.cam_utils import prepare_camera_embedding
-
-# from fastvideo.configs.sample import SamplingParam
-OUTPUT_PATH = "video_samples_lingbotworld"
-def main():
-    # FastVideo will automatically use the optimal default arguments for the
-    # model.
-    # If a local path is provided, FastVideo will make a best effort
-    # attempt to identify the optimal arguments.
-    generator = VideoGenerator.from_pretrained(
-         "FastVideo/LingBot-World-Base-Cam-Diffusers",
-        # FastVideo will automatically handle distributed setup
-        num_gpus=1,
-        use_fsdp_inference=False, # set to True if GPU is out of memory
-        dit_cpu_offload=True, # DiT need to be offloaded for MoE
-        vae_cpu_offload=False,
-        text_encoder_cpu_offload=True,
-        # Set pin_cpu_memory to false if CPU RAM is limited and there're no frequent CPU-GPU transfer
-        pin_cpu_memory=True,
-        # image_encoder_cpu_offload=False,
-    )
-
-    num_frames = 81
-    prompt = "The video presents a soaring journey through a fantasy jungle. The wind whips past the rider's blue hands gripping the reins, causing the leather straps to vibrate. The ancient gothic castle approaches steadily, its stone details becoming clearer against the backdrop of floating islands and distant waterfalls."
-    image_path = "https://raw.githubusercontent.com/Robbyant/lingbot-world/main/examples/00/image.jpg"
-    action_path = "examples/inference/basic/lingbotworld_examples/00"
-    c2ws_plucker_emb, num_frames = prepare_camera_embedding(
-        action_path=action_path,
-        num_frames=num_frames,
-        height=480,
-        width=832,
-        spatial_scale=8,
-    )
-
-    generator.generate_video(
-        prompt,
-        image_path=image_path,
-        output_path=OUTPUT_PATH,
-        save_video=True,
-        num_frames=num_frames,
-        height=480,
-        width=832,
-        c2ws_plucker_emb=c2ws_plucker_emb,
-    )
-
-
-if __name__ == "__main__":
-    main()
@@ -1,6 +1,5 @@
 from fastvideo import VideoGenerator

-
 PROMPT = (
    "A warm sunny backyard. The camera starts in a tight cinematic close-up "
    "of a woman and a man in their 30s, facing each other with serious "
@@ -17,26 +16,16 @@ PROMPT = (


 def main() -> None:
-    # Uses FastVideo default sampling settings for LTX2 base.
    generator = VideoGenerator.from_pretrained(
-        "Davids048/LTX2-Base-Diffusers",
-        num_gpus=8,
+        "FastVideo/LTX2-Distilled-Diffusers",
+        num_gpus=1,
    )

-    output_path = "outputs_video/ltx2_basic/output_ltx2_base_t2v_1088_1920_1.4.mp4"
+    output_path = "outputs_video/ltx2_basic/output_ltx2_distilled_t2v.mp4"
    generator.generate_video(
        prompt=PROMPT,
        output_path=output_path,
        save_video=True,
-        num_frames=121,
-        height=1088,
-        width=1920,
-        # LTX2 uses these parameters for multi-modal CFG instead of guidance_scale
-        # ltx2_cfg_scale_video=3.0,
-        # ltx2_cfg_scale_audio=7.0,
-        # ltx2_modality_scale_video=3.0,
-        # ltx2_modality_scale_audio=3.0,
-        # ltx2_rescale_scale=0.7,
    )
    generator.shutdown()

@@ -1,34 +0,0 @@
-from fastvideo import VideoGenerator
-
-PROMPT = (
-    "A warm sunny backyard. The camera starts in a tight cinematic close-up "
-    "of a woman and a man in their 30s, facing each other with serious "
-    "expressions. The woman, emotional and dramatic, says softly, \"That's "
-    "it... Dad's lost it. And we've lost Dad.\" The man exhales, slightly "
-    "annoyed: \"Stop being so dramatic, Jess.\" A beat. He glances aside, "
-    "then mutters defensively, \"He's just having fun.\" The camera slowly "
-    "pans right, revealing the grandfather in the garden wearing enormous "
-    "butterfly wings, waving his arms in the air like he's trying to take "
-    "off. He shouts, \"Wheeeew!\" as he flaps his wings with full commitment. "
-    "The woman covers her face, on the verge of tears. The tone is deadpan, "
-    "absurd, and quietly tragic."
-)
-
-
-def main() -> None:
-    generator = VideoGenerator.from_pretrained(
-        "FastVideo/LTX2-Distilled-Diffusers",
-        num_gpus=1,
-    )
-
-    output_path = "outputs_video/ltx2_basic/output_ltx2_distilled_t2v.mp4"
-    generator.generate_video(
-        prompt=PROMPT,
-        output_path=output_path,
-        save_video=True,
-    )
-    generator.shutdown()
-
-
-if __name__ == "__main__":
-    main()
@@ -1,137 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-
-from __future__ import annotations
-
-import argparse
-import os
-import re
-from typing import List
-
-
-DEFAULT_PROMPTS = [
-    "a photo of a cat",
-    "a cinematic photo of a red panda wearing a tiny backpack, standing on a rainy neon-lit street at night, shallow depth of field, sharp focus, 35mm, bokeh",
-]
-
-
-def _safe_filename(text: str, max_len: int = 100) -> str:
-    """
-    Make a stable, filesystem-friendly filename base.
-    VideoGenerator uses prompt[:100].strip() internally, so we mirror that,
-    but also remove path separators and other problematic characters.
-    """
-    s = text[:max_len].strip()
-    s = s.replace(os.sep, "_")
-    if os.altsep:
-        s = s.replace(os.altsep, "_")
-    s = re.sub(r"\s+", " ", s)
-    s = re.sub(r"[^A-Za-z0-9 .,_-]", "_", s)
-    s = s.strip(" .")
-    return s or "prompt"
-
-
-def _remove_existing_outputs(out_dir: str, filename_base: str) -> None:
-    """
-    Ensure deterministic naming by deleting any existing mp4s that would
-    cause VideoGenerator to append suffixes like _1, _2, etc.
-    """
-    if not os.path.isdir(out_dir):
-        return
-
-    pattern = re.compile(rf"^{re.escape(filename_base)}(_\d+)?\.mp4$")
-    for fn in os.listdir(out_dir):
-        if pattern.match(fn):
-            try:
-                os.remove(os.path.join(out_dir, fn))
-            except FileNotFoundError:
-                pass
-
-
-def parse_args() -> argparse.Namespace:
-    p = argparse.ArgumentParser(description="Run SD3.5 Medium text-to-image with FastVideo VideoGenerator.")
-    p.add_argument("--model-path", default="stabilityai/stable-diffusion-3.5-medium", help="Path to local diffusers-format SD3.5 weights directory.")
-    p.add_argument(
-        "--out-dir",
-        "--outdir",
-        default="outputs/sd35/samples",
-        help="Output directory for generated mp4 files.",
-    )
-    p.add_argument(
-        "--prompt",
-        action="append",
-        default=None,
-        help="Prompt text. Repeat --prompt multiple times to generate multiple samples.",
-    )
-    p.add_argument("--negative", default="lowres, blurry, jpeg artifacts, watermark, text", help="Negative prompt.")
-    p.add_argument(
-        "--backend",
-        default=None,
-        help="Set FASTVIDEO_ATTENTION_BACKEND (e.g. TORCH_SDPA). If omitted, respects the existing env var.",
-    )
-    p.add_argument("--seed", type=int, default=42, help="Base seed. Each prompt uses seed + prompt_idx.")
-    p.add_argument("--height", type=int, default=768, help="Output height.")
-    p.add_argument("--width", type=int, default=768, help="Output width.")
-    p.add_argument("--steps", type=int, default=28, help="Number of inference steps.")
-    p.add_argument("--guidance", type=float, default=6.0, help="Guidance scale.")
-    p.add_argument("--num-gpus", type=int, default=1, help="Number of GPUs to use.")
-    return p.parse_args()
-
-
-def main() -> None:
-    args = parse_args()
-    prompts: List[str] = args.prompt if args.prompt else DEFAULT_PROMPTS
-
-    if args.backend:
-        os.environ["FASTVIDEO_ATTENTION_BACKEND"] = args.backend
-
-    from fastvideo import VideoGenerator
-
-    os.makedirs(args.out_dir, exist_ok=True)
-
-    init_kwargs = {
-        "num_gpus": args.num_gpus,
-        "workload_type": "t2i",
-        "sp_size": 1,
-        "tp_size": 1,
-        "dit_cpu_offload": False,
-        "dit_layerwise_offload": False,
-        "text_encoder_cpu_offload": False,
-        "vae_cpu_offload": False,
-        "image_encoder_cpu_offload": False,
-        "pin_cpu_memory": False,
-        "use_fsdp_inference": False,
-    }
-
-    generator = VideoGenerator.from_pretrained(model_path=args.model_path, **init_kwargs)
-    try:
-        for i, prompt in enumerate(prompts):
-            seed = args.seed + i
-
-            filename_base = f"sd35_{i:02d}_seed{seed}_{_safe_filename(prompt, max_len=80)}"
-            _remove_existing_outputs(args.out_dir, filename_base)
-
-            output_path = os.path.join(args.out_dir, f"{filename_base}.png")
-            print(f"[sd35] prompt_idx={i} seed={seed} output_path={output_path}")
-
-            generation_kwargs = {
-                "output_path": output_path,
-                "height": args.height,
-                "width": args.width,
-                "num_frames": 1,
-                "fps": 1,
-                "num_inference_steps": args.steps,
-                "guidance_scale": args.guidance,
-                "seed": seed,
-                "negative_prompt": args.negative,
-                "save_video": True,
-            }
-
-            generator.generate_video(prompt, **generation_kwargs)
-
-        print(f"[sd35] done. outputs written to: {args.out_dir}")
-    finally:
-        generator.shutdown()
-
-
-if __name__ == "__main__":
-    main()
@@ -31,7 +31,7 @@ def main():
    sampling_param.height = 480
    sampling_param.seed = 1000

-    with open("assets/prompts/mixkit_i2v.jsonl", "r") as f:
+    with open("prompts/mixkit_i2v.jsonl", "r") as f:
        prompt_image_pairs = json.load(f)

    for prompt_image_pair in prompt_image_pairs:
@@ -10,12 +10,13 @@ def main():
    # If a local path is provided, FastVideo will make a best effort
    # attempt to identify the optimal arguments.
    generator = VideoGenerator.from_pretrained(
-        "IRMChen/Wan2.1-Fun-1.3B-Control-Diffusers",
+        "weizhou03/Wan2.1-Fun-1.3B-InP-Diffusers",
+        # "IRMChen/Wan2.1-Fun-1.3B-Control-Diffusers",
        # "alibaba-pai/Wan2.2-Fun-A14B-Control",
        # FastVideo will automatically handle distributed setup
        num_gpus=1,
        use_fsdp_inference=False, # set to True if GPU is out of memory
-        dit_cpu_offload=True, # DiT need to be offloaded for MoE
+        dit_cpu_offload=False, # DiT need to be offloaded for MoE
        vae_cpu_offload=False,
        text_encoder_cpu_offload=True,
        # Set pin_cpu_memory to false if CPU RAM is limited and there're no frequent CPU-GPU transfer
@@ -23,14 +24,19 @@ def main():
        # image_encoder_cpu_offload=False,
    )

-    prompt = "一位年轻女性穿着一件粉色的连衣裙，裙子上有白色的装饰和粉色的纽扣。她的头发是紫色的，头上戴着一个红色的大蝴蝶结，显得非常可爱和精致。她还戴着一个红色的领结，整体造型充满了少女感和活力。她的表情温柔，双手轻轻交叉放在身前，姿态优雅。背景是简单的灰色，没有任何多余的装饰，使得人物更加突出。她的妆容清淡自然，突显了她的清新气质。整体画面给人一种甜美、梦幻的感觉，仿佛置身于童话世界中。"
-    negative_prompt = "色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走"
+    # prompt = "一位年轻女性穿着一件粉色的连衣裙，裙子上有白色的装饰和粉色的纽扣。她的头发是紫色的，头上戴着一个红色的大蝴蝶结，显得非常可爱和精致。她还戴着一个红色的领结，整体造型充满了少女感和活力。她的表情温柔，双手轻轻交叉放在身前，姿态优雅。背景是简单的灰色，没有任何多余的装饰，使得人物更加突出。她的妆容清淡自然，突显了她的清新气质。整体画面给人一种甜美、梦幻的感觉，仿佛置身于童话世界中。"
+    # negative_prompt = "色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走"
    # prompt                  = "A young woman with beautiful, clear eyes and blonde hair stands in the forest, wearing a white dress and a crown. Her expression is serene, reminiscent of a movie star, with fair and youthful skin. Her brown long hair flows in the wind. The video quality is very high, with a clear view. High quality, masterpiece, best quality, high resolution, ultra-fine, fantastical."
    # negative_prompt         = "Twisted body, limb deformities, text captions, comic, static, ugly, error, messy code."
-    image_path = "https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/8.png"
-    control_video_path = "https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/pose.mp4"
-
-    video = generator.generate_video(prompt, negative_prompt=negative_prompt, image_path=image_path, video_path=control_video_path, output_path=OUTPUT_PATH, output_video_name=OUTPUT_NAME, save_video=True)
+    # image_path = "https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/8.png"
+    # control_video_path = "https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/pose.mp4"
+    import json, os
+    with open("/mnt/weka/home/hao.zhang/wei/FastVideo/data/mixkit_i2v.jsonl", "r") as f:
+        prompt_image_pairs = json.load(f)
+    for d in prompt_image_pairs:
+        prompt = d["prompt"]
+        image_path = os.path.join("/mnt/weka/home/hao.zhang/wei/FastVideo", d["image_path"])
+        video = generator.generate_video(prompt, image_path=image_path, output_path=OUTPUT_PATH, output_video_name=OUTPUT_NAME, save_video=True)

 if __name__ == "__main__":
    main()
@@ -1,6 +1,6 @@
 from fastvideo import VideoGenerator

-OUTPUT_PATH = "video_samples_wan2_2_5B_ti2v"
+OUTPUT_PATH = "video_samples_wan2_2_5B_ti2v_720p_77"
 def main():
    # FastVideo will automatically use the optimal default arguments for the
    # model.
@@ -12,17 +12,29 @@ def main():
        # FastVideo will automatically handle distributed setup
        num_gpus=1,
        use_fsdp_inference=False, # set to True if GPU is out of memory
-        dit_cpu_offload=True,
+        dit_cpu_offload=False,
        vae_cpu_offload=False,
        text_encoder_cpu_offload=True,
        pin_cpu_memory=True, # set to false if low CPU RAM or hit obscure "CUDA error: Invalid argument"
        # image_encoder_cpu_offload=False,
    )

+    import json, os
+    with open("/mnt/weka/home/hao.zhang/wei/FastVideo/data/mixkit_i2v_720p.jsonl", "r") as f:
+        prompt_image_pairs = json.load(f)
+    for d in prompt_image_pairs:
+        prompt = d["prompt"]
+        image_path = os.path.join("/mnt/weka/home/hao.zhang/wei/FastVideo", d["image_path"])
+        video = generator.generate_video(prompt, image_path=image_path, output_path=OUTPUT_PATH, output_video_name=OUTPUT_PATH, save_video=True, num_frames=77)
+    return
+
    # I2V is triggered just by passing in an image_path argument
    prompt = "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside."
    image_path = "https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/wan_i2v_input.JPG"
+    # prompt = "In the midst of the joyous New Year's Eve celebration, the cheerful group of friends, their spirits lifted by the festivities, decides to immortalize the moment with a vibrant snapshot"
+    # image_path = "images/mixkit-a-cheerful-group-of-friends-celebrate-new-years-eve-and-51525.png"
    video = generator.generate_video(prompt, output_path=OUTPUT_PATH, save_video=True, image_path=image_path)
+    return

    # Generate another video with a different prompt, without reloading the
    # model!
@@ -188,9 +188,8 @@ def load_example_prompts():
    prompt_to_image = {}
    # Try to find the JSON file relative to project root
    possible_json_paths = [
-        Path("assets/prompts/mixkit_i2v.jsonl"),
-        Path(__file__).resolve().parents[4] / "assets" / "prompts" /
-        "mixkit_i2v.jsonl",
+        Path("prompts/mixkit_i2v.jsonl"),
+        Path(__file__).parent.parent.parent.parent / "prompts" / "mixkit_i2v.jsonl",
    ]
    json_path = None
    for path in possible_json_paths:
@@ -202,8 +201,8 @@ def load_example_prompts():
        try:
            with open(json_path, "r", encoding='utf-8') as f:
                data = json.load(f)
-                # Resolve paths relative to repository root.
-                project_root = Path(__file__).resolve().parents[4]
+                # Get the project root directory (parent of prompts directory)
+                project_root = json_path.parent.parent
                for item in data:
                    prompt_text = item.get("prompt", "").strip()
                    image_path = item.get("image_path", "")
@@ -737,8 +736,8 @@ def main():
        allowed_paths=[
            os.path.abspath("outputs"), 
            os.path.abspath("fastvideo-logos"),
-            os.path.abspath("assets/prompts"),
-            os.path.abspath("assets/images"),
+            os.path.abspath("prompts"),
+            os.path.abspath("images"),
            os.path.abspath(tempfile.gettempdir()),
            os.path.abspath(os.path.join(tempfile.gettempdir(), "gradio")),
        ]
@@ -748,4 +747,4 @@ def main():


 if __name__ == "__main__":
-    main() 
+    main() 
@@ -4,7 +4,7 @@ export WANDB_BASE_URL="https://api.wandb.ai"
 export WANDB_MODE=online
 export TOKENIZERS_PARALLELISM=false

-MODEL_PATH="wlsaidhi/SFWan2.1-T2V-1.3B-Diffusers"
+MODEL_PATH="Wan-AI/Wan2.1-T2V-1.3B-Diffusers"
 DATA_DIR="data/crush-smol_processed_t2v_1_3b_ode_init/"
 VALIDATION_DATASET_FILE="$(dirname "$0")/validation.json"
 NUM_GPUS=1
@@ -14,6 +14,7 @@ NUM_GPUS=1
 training_args=(
  --tracker_project_name "wan_ode_init"
  --output_dir "wan_ode_init_crush_smol"
+  --override_transformer_cls_name "CausalWanTransformer3DModel"
  --wandb_run_name "wan_ode_init_crush_smol"
  --max_train_steps 6000
  --train_batch_size 1
@@ -33,7 +34,7 @@ parallel_args=(
  --sp_size 1
  --tp_size 1
  --hsdp_replicate_dim 1
-  --hsdp_shard_dim $NUM_GPUS
+  --hsdp_shard_dim 1
 )

 # Model arguments
@@ -50,17 +51,20 @@ dataset_args=(

 # Validation arguments
 validation_args=(
-  --log-visualization
-  --visualization-steps 100
+  --log_validation
+  --validation_dataset_file "$VALIDATION_DATASET_FILE"
+  --validation_steps 50
+  --validation_sampling_steps "50"
+  --validation_guidance_scale "6.0"
 )

 # Optimizer arguments
 optimizer_args=(
-  --learning_rate 1e-5
+  --learning_rate 6e-6
  --mixed_precision "bf16"
  --weight_only_checkpointing_steps 1000
  --training_state_checkpointing_steps 1000
-  --weight_decay 0.01
+  --weight_decay 1e-4
  --max_grad_norm 1.0
 )

@@ -0,0 +1,124 @@
+#!/bin/bash
+#SBATCH --job-name=hy15_ode_3333
+#SBATCH --partition=main
+#SBATCH --nodes=8
+#SBATCH --ntasks=8
+#SBATCH --ntasks-per-node=1
+#SBATCH --gres=gpu:8
+#SBATCH --cpus-per-task=128
+#SBATCH --mem=1440G
+#SBATCH --output=hy15_ode_3333_output/hy15_ode_3333_%j.out
+#SBATCH --error=hy15_ode_3333_output/hy15_ode_3333_%j.err
+#SBATCH --exclusive
+
+# Basic Info
+export NCCL_P2P_DISABLE=1
+export TORCH_NCCL_ENABLE_MONITORING=0
+export NCCL_DEBUG_SUBSYS=INIT,NET
+# different cache dir for different processes
+export TRITON_CACHE_DIR=/tmp/triton_cache_${SLURM_PROCID}
+export MASTER_PORT=29500
+export NODE_RANK=$SLURM_PROCID
+nodes=( $(scontrol show hostnames $SLURM_JOB_NODELIST) )
+export MASTER_ADDR=${nodes[0]}
+export TOKENIZERS_PARALLELISM=false
+export WANDB_API_KEY='your_wandb_api_key_here'
+export WANDB_BASE_URL="https://api.wandb.ai"
+export WANDB_MODE=online
+export FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN
+
+# Configs
+NUM_GPUS=8
+NUM_TOTAL_GPUS=64
+
+MODEL_PATH="weizhou03/SFHunyuanVideo-1.5-Diffusers-480p_t2v"
+DATA_DIR="your_data_dir"
+VALIDATION_DATASET_FILE="your_validation_data_dir"
+# IP=[MASTER NODE IP]
+
+# Training arguments
+training_args=(
+  --tracker_project_name "hy15_ode_init"
+  --output_dir "your_output_dir"
+  --wandb_run_name "your_wandb_run_name"
+  --warp_denoising_step
+  --dmd_denoising_steps '1000,760,520,280,0'
+  --log_visualization
+  --visualization_steps 100
+  --max_train_steps 4000
+  --train_batch_size 1
+  --train_sp_batch_size 1
+  --gradient_accumulation_steps 2
+  --num_latent_t 21
+  --num_height 480
+  --num_width 848
+  --num_frames 81
+  --enable_gradient_checkpointing_type "full"
+)
+
+# Parallel arguments
+parallel_args=(
+  --num_gpus $NUM_TOTAL_GPUS # 64
+  --sp_size 1
+  --tp_size 1
+  --hsdp_replicate_dim 1 # 64
+  --hsdp_shard_dim $NUM_TOTAL_GPUS
+)
+
+# Model arguments
+model_args=(
+  --model_path $MODEL_PATH
+  --pretrained_model_name_or_path $MODEL_PATH
+  --text-encoder-cpu-offload
+)
+
+# Dataset arguments
+dataset_args=(
+  --data_path "$DATA_DIR" 
+  --dataloader_num_workers 4
+)
+
+# Optimizer arguments
+optimizer_args=(
+  --learning_rate 1e-5
+  --mixed_precision "bf16"
+  --training_state_checkpointing_steps 400
+  --weight_decay 0.01
+  --max_grad_norm 1.0
+)
+
+validation_args=(
+  # --log_validation
+  --validation_dataset_file "$VALIDATION_DATASET_FILE"
+  --validation_steps 10
+  --validation_sampling_steps "4"
+  --validation_guidance_scale "1.0" # not used for dmd inference
+  # --vae_cpu_offload True
+)
+
+# Miscellaneous arguments
+miscellaneous_args=(
+  --inference_mode False
+  --checkpoints_total_limit 6
+  --training_cfg_rate 0.1
+  --multi_phased_distill_schedule "4000-1"
+  --not_apply_cfg_solver
+  --dit_precision "fp32"
+  --num_euler_timesteps 50
+)
+
+# If you do not have 32 GPUs and to fit in memory, you can: 1. increase sp_size. 2. reduce num_latent_t
+srun torchrun \
+--nnodes $SLURM_JOB_NUM_NODES \
+--nproc_per_node $NUM_GPUS \
+--node_rank $SLURM_PROCID \
+--rdzv_backend=c10d \
+--rdzv_endpoint="$MASTER_ADDR:$MASTER_PORT" \
+    fastvideo/training/ode_causal_pipeline.py \
+    "${parallel_args[@]}" \
+    "${model_args[@]}" \
+    "${dataset_args[@]}" \
+    "${training_args[@]}" \
+    "${optimizer_args[@]}" \
+    "${validation_args[@]}" \
+    "${miscellaneous_args[@]}"
@@ -0,0 +1,133 @@
+#!/bin/bash
+#SBATCH --job-name=hy15_tf_3333
+#SBATCH --partition=main
+#SBATCH --nodes=8
+#SBATCH --ntasks=8
+#SBATCH --ntasks-per-node=1
+#SBATCH --gres=gpu:8
+#SBATCH --cpus-per-task=128
+#SBATCH --mem=1440G
+#SBATCH --output=hy15_tf_3333_output/hy15_tf_3333_%j.out
+#SBATCH --error=hy15_tf_3333_output/hy15_tf_3333_%j.err
+#SBATCH --exclusive
+
+# Basic Info
+export NCCL_P2P_DISABLE=1
+export TORCH_NCCL_ENABLE_MONITORING=0
+export NCCL_DEBUG_SUBSYS=INIT,NET
+# different cache dir for different processes
+export TRITON_CACHE_DIR=/tmp/triton_cache_${SLURM_PROCID}
+export MASTER_PORT=29500
+export NODE_RANK=$SLURM_PROCID
+nodes=( $(scontrol show hostnames $SLURM_JOB_NODELIST) )
+export MASTER_ADDR=${nodes[0]}
+export TOKENIZERS_PARALLELISM=false
+# export WANDB_API_KEY="8d9f4b39abd68eb4e29f6fc010b7ee71a2207cde"
+export WANDB_API_KEY="50632ebd88ffd970521cec9ab4a1a2d7e85bfc45"
+# export WANDB_API_KEY='your_wandb_api_key_here'
+export WANDB_BASE_URL="https://api.wandb.ai"
+export WANDB_MODE=online
+export FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN
+
+source ~/conda/miniconda/bin/activate
+conda activate wei-fv-distill
+export HOME="/mnt/weka/home/hao.zhang/wei"
+
+# Configs
+NUM_GPUS=8
+NUM_TOTAL_GPUS=64
+
+MODEL_PATH="weizhou03/SFHunyuanVideo-1.5-Diffusers-480p_t2v"
+DATA_DIR="/mnt/weka/home/hao.zhang/wl/sharefs/wl/release/fv-ode-preprocessing-16k-hy15-121"
+VALIDATION_DATASET_FILE="/mnt/weka/home/hao.zhang/wl/FastVideo/examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free/validation_64.json"
+# IP=[MASTER NODE IP]
+
+# Training arguments
+training_args=(
+  --tracker_project_name "hy15_ode_init"
+  --output_dir "/mnt/weka/home/hao.zhang/wei/hy15_worldplay_tf_init_3333"
+  --wandb_run_name "hy15_worldplay_tf_init_3333"
+  --warp_denoising_step
+  --dmd_denoising_steps '1000,760,520,280,0'
+  --log_visualization
+  --visualization_steps 100
+  --max_train_steps 30000
+  --train_batch_size 1
+  --train_sp_batch_size 1
+  --gradient_accumulation_steps 1
+  --num_latent_t 21
+  --num_height 480
+  --num_width 848
+  --num_frames 81
+  --enable_gradient_checkpointing_type "full"
+)
+
+# Parallel arguments
+parallel_args=(
+  --num_gpus $NUM_TOTAL_GPUS # 64
+  --sp_size 1
+  --tp_size 1
+  --hsdp_replicate_dim 1 # 64
+  --hsdp_shard_dim $NUM_TOTAL_GPUS
+)
+
+# Model arguments
+model_args=(
+  --model_path $MODEL_PATH
+  --pretrained_model_name_or_path $MODEL_PATH
+  --text-encoder-cpu-offload
+)
+
+# Dataset arguments
+dataset_args=(
+  --data_path "$DATA_DIR" 
+  --dataloader_num_workers 4
+)
+
+# Optimizer arguments
+optimizer_args=(
+  --optimizer-type "muon"
+  --learning_rate 1e-5
+  --mixed_precision "bf16"
+  --training_state_checkpointing_steps 400
+  --weight_decay 0.01
+  --max_grad_norm 1.0
+)
+
+validation_args=(
+  # --log_validation
+  --validation_dataset_file "$VALIDATION_DATASET_FILE"
+  --validation_steps 10
+  --validation_sampling_steps "4"
+  --validation_guidance_scale "1.0" # not used for dmd inference
+  # --vae_cpu_offload True
+)
+
+# Miscellaneous arguments
+miscellaneous_args=(
+  --inference_mode False
+  --checkpoints_total_limit 6
+  --training_cfg_rate 0.1
+  --multi_phased_distill_schedule "4000-1"
+  --not_apply_cfg_solver
+  --dit_precision "fp32"
+  --num_euler_timesteps 50
+  --ema_start_step 0
+  # --enable_gradient_checkpointing_type "full"
+)
+
+# If you do not have 32 GPUs and to fit in memory, you can: 1. increase sp_size. 2. reduce num_latent_t
+srun torchrun \
+--nnodes $SLURM_JOB_NUM_NODES \
+--nproc_per_node $NUM_GPUS \
+--node_rank $SLURM_PROCID \
+--rdzv_backend=c10d \
+--rdzv_endpoint="$MASTER_ADDR:$MASTER_PORT" \
+    fastvideo/training/hy15_ode_causal_pipeline.py \
+    "${parallel_args[@]}" \
+    "${model_args[@]}" \
+    "${dataset_args[@]}" \
+    "${training_args[@]}" \
+    "${optimizer_args[@]}" \
+    "${validation_args[@]}" \
+    "${miscellaneous_args[@]}"
@@ -0,0 +1,137 @@
+#!/bin/bash
+#SBATCH --job-name=wan_tf_3333
+#SBATCH --nodes=8
+#SBATCH --ntasks=8
+#SBATCH --ntasks-per-node=1
+#SBATCH --gres=gpu:8
+#SBATCH --cpus-per-task=128
+#SBATCH --mem=1440G
+#SBATCH --output=wan_tf_3333_output/wan_tf_3333_%j.out
+#SBATCH --error=wan_tf_3333_output/wan_tf_3333_%j.err
+#SBATCH --exclusive
+
+# Basic Info
+export NCCL_P2P_DISABLE=1
+export TORCH_NCCL_ENABLE_MONITORING=0
+export NCCL_DEBUG_SUBSYS=INIT,NET
+# different cache dir for different processes
+# export TRITON_CACHE_DIR=/tmp/triton_cache_${SLURM_PROCID}
+export MASTER_PORT=29500
+export NODE_RANK=$SLURM_PROCID
+nodes=( $(scontrol show hostnames $SLURM_JOB_NODELIST) )
+export MASTER_ADDR=${nodes[0]}
+export TOKENIZERS_PARALLELISM=false
+# export WANDB_API_KEY="8d9f4b39abd68eb4e29f6fc010b7ee71a2207cde"
+export WANDB_API_KEY="50632ebd88ffd970521cec9ab4a1a2d7e85bfc45"
+# export WANDB_API_KEY='your_wandb_api_key_here'
+export WANDB_BASE_URL="https://api.wandb.ai"
+export WANDB_MODE=online
+export FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN
+
+export HF_HOME=/mnt/data/huggingface
+export CUDA_HOME=/usr/local/cuda
+export PATH="$CUDA_HOME/bin:$PATH"
+export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/home/shared-bin/local/usr/lib/aarch64-linux-gnu:$CUDA_HOME/lib64"
+export CUDACXX="$CUDA_HOME/bin/nvcc"
+
+source ~/miniconda3/bin/activate
+conda activate fastvideo
+
+# Configs
+NUM_GPUS=8
+NUM_TOTAL_GPUS=64
+
+MODEL_PATH="Wan-AI/Wan2.1-T2V-1.3B-Diffusers"
+DATA_DIR="/mnt/data/mixkit_wan_1.3b_processed_t2v_distributed"
+VALIDATION_DATASET_FILE="/mnt/weka/home/hao.zhang/wl/FastVideo/examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free/validation_64.json"
+# IP=[MASTER NODE IP]
+
+# Training arguments
+training_args=(
+  --tracker_project_name "wan_tf_init"
+  --output_dir "/mnt/data/wan_ar_df_diffusion_3333_mixkit_shift5"
+  --wandb_run_name "wan_df_init_3333"
+  --warp_denoising_step
+  --dmd_denoising_steps '1000,750,500,250,0'
+  --log_visualization
+  --visualization_steps 100
+  --max_train_steps 30000
+  --train_batch_size 1
+  --train_sp_batch_size 1
+  --gradient_accumulation_steps 1
+  --num_latent_t 21
+  --num_height 480
+  --num_width 832
+  --num_frames 81
+  --override-transformer-cls-name "CausalWanTransformer3DModel"
+)
+
+# Parallel arguments
+parallel_args=(
+  --num_gpus $NUM_TOTAL_GPUS # 64
+  --sp_size 1
+  --tp_size 1
+  --hsdp_replicate_dim 1 # 64
+  --hsdp_shard_dim $NUM_TOTAL_GPUS
+)
+
+# Model arguments
+model_args=(
+  --model_path $MODEL_PATH
+  --pretrained_model_name_or_path $MODEL_PATH
+  --text-encoder-cpu-offload
+)
+
+# Dataset arguments
+dataset_args=(
+  --data_path "$DATA_DIR" 
+  --dataloader_num_workers 4
+)
+
+# Optimizer arguments
+optimizer_args=(
+  --optimizer-type "adamw"
+  --learning_rate 1e-5
+  --mixed_precision "bf16"
+  --training_state_checkpointing_steps 800
+  --weight_decay 0.01
+  --max_grad_norm 1.0
+)
+
+validation_args=(
+  # --log_validation
+  --validation_dataset_file "$VALIDATION_DATASET_FILE"
+  --validation_steps 10
+  --validation_sampling_steps "4"
+  --validation_guidance_scale "1.0" # not used for dmd inference
+  # --vae_cpu_offload True
+)
+
+# Miscellaneous arguments
+miscellaneous_args=(
+  --inference_mode False
+  --checkpoints_total_limit 6
+  --training_cfg_rate 0.1
+  --multi_phased_distill_schedule "4000-1"
+  --not_apply_cfg_solver
+  --dit_precision "fp32"
+  --num_euler_timesteps 50
+  --ema_start_step 0
+  --enable_gradient_checkpointing_type "full"
+)
+
+# If you do not have 32 GPUs and to fit in memory, you can: 1. increase sp_size. 2. reduce num_latent_t
+srun torchrun \
+--nnodes $SLURM_JOB_NUM_NODES \
+--nproc_per_node $NUM_GPUS \
+--node_rank $SLURM_PROCID \
+--rdzv_backend=c10d \
+--rdzv_endpoint="$MASTER_ADDR:$MASTER_PORT" \
+    fastvideo/training/ar_diffusion_pipeline.py \
+    "${parallel_args[@]}" \
+    "${model_args[@]}" \
+    "${dataset_args[@]}" \
+    "${training_args[@]}" \
+    "${optimizer_args[@]}" \
+    "${validation_args[@]}" \
+    "${miscellaneous_args[@]}"
@@ -0,0 +1,138 @@
+#!/bin/bash
+#SBATCH --job-name=wan_tf_3333
+#SBATCH --nodes=8
+#SBATCH --ntasks=8
+#SBATCH --ntasks-per-node=1
+#SBATCH --gres=gpu:8
+#SBATCH --cpus-per-task=128
+#SBATCH --mem=1440G
+#SBATCH --output=wan_tf_3333_output/wan_tf_3333_%j.out
+#SBATCH --error=wan_tf_3333_output/wan_tf_3333_%j.err
+#SBATCH --exclusive
+
+# Basic Info
+export NCCL_P2P_DISABLE=1
+export TORCH_NCCL_ENABLE_MONITORING=0
+export NCCL_DEBUG_SUBSYS=INIT,NET
+# different cache dir for different processes
+# export TRITON_CACHE_DIR=/tmp/triton_cache_${SLURM_PROCID}
+export MASTER_PORT=29500
+export NODE_RANK=$SLURM_PROCID
+nodes=( $(scontrol show hostnames $SLURM_JOB_NODELIST) )
+export MASTER_ADDR=${nodes[0]}
+export TOKENIZERS_PARALLELISM=false
+# export WANDB_API_KEY="8d9f4b39abd68eb4e29f6fc010b7ee71a2207cde"
+export WANDB_API_KEY="50632ebd88ffd970521cec9ab4a1a2d7e85bfc45"
+# export WANDB_API_KEY='your_wandb_api_key_here'
+export WANDB_BASE_URL="https://api.wandb.ai"
+export WANDB_MODE=online
+export FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN
+
+export HF_HOME=/mnt/data/huggingface
+export CUDA_HOME=/usr/local/cuda
+export PATH="$CUDA_HOME/bin:$PATH"
+export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/home/shared-bin/local/usr/lib/aarch64-linux-gnu:$CUDA_HOME/lib64"
+export CUDACXX="$CUDA_HOME/bin/nvcc"
+
+source ~/miniconda3/bin/activate
+conda activate fastvideo
+
+# Configs
+NUM_GPUS=8
+NUM_TOTAL_GPUS=64
+
+MODEL_PATH="Wan-AI/Wan2.1-T2V-1.3B-Diffusers"
+DATA_DIR="/mnt/data/mixkit_wan_1.3b_processed_t2v_distributed"
+VALIDATION_DATASET_FILE="/mnt/weka/home/hao.zhang/wl/FastVideo/examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free/validation_64.json"
+# IP=[MASTER NODE IP]
+
+# Training arguments
+training_args=(
+  --tracker_project_name "wan_tf_init"
+  --output_dir "/mnt/data/wan_ar_diffusion_3333_mixkit_shift5"
+  --wandb_run_name "wan_tf_init_3333"
+  --warp_denoising_step
+  --dmd_denoising_steps '1000,750,500,250,0'
+  --log_visualization
+  --visualization_steps 100
+  --max_train_steps 30000
+  --train_batch_size 1
+  --train_sp_batch_size 1
+  --gradient_accumulation_steps 1
+  --num_latent_t 21
+  --num_height 480
+  --num_width 832
+  --num_frames 81
+  --override-transformer-cls-name "CausalWanTransformer3DModel"
+)
+
+# Parallel arguments
+parallel_args=(
+  --num_gpus $NUM_TOTAL_GPUS # 64
+  --sp_size 1
+  --tp_size 1
+  --hsdp_replicate_dim 1 # 64
+  --hsdp_shard_dim $NUM_TOTAL_GPUS
+)
+
+# Model arguments
+model_args=(
+  --model_path $MODEL_PATH
+  --pretrained_model_name_or_path $MODEL_PATH
+  --text-encoder-cpu-offload
+)
+
+# Dataset arguments
+dataset_args=(
+  --data_path "$DATA_DIR" 
+  --dataloader_num_workers 4
+)
+
+# Optimizer arguments
+optimizer_args=(
+  --optimizer-type "adamw"
+  --learning_rate 1e-5
+  --mixed_precision "bf16"
+  --training_state_checkpointing_steps 800
+  --weight_decay 0.01
+  --max_grad_norm 1.0
+)
+
+validation_args=(
+  # --log_validation
+  --validation_dataset_file "$VALIDATION_DATASET_FILE"
+  --validation_steps 10
+  --validation_sampling_steps "4"
+  --validation_guidance_scale "1.0" # not used for dmd inference
+  # --vae_cpu_offload True
+)
+
+# Miscellaneous arguments
+miscellaneous_args=(
+  --inference_mode False
+  --checkpoints_total_limit 6
+  --training_cfg_rate 0.1
+  --multi_phased_distill_schedule "4000-1"
+  --not_apply_cfg_solver
+  --dit_precision "fp32"
+  --num_euler_timesteps 50
+  --ema_start_step 0
+  --enable_gradient_checkpointing_type "full"
+  --use-tf True
+)
+
+# If you do not have 32 GPUs and to fit in memory, you can: 1. increase sp_size. 2. reduce num_latent_t
+srun torchrun \
+--nnodes $SLURM_JOB_NUM_NODES \
+--nproc_per_node $NUM_GPUS \
+--node_rank $SLURM_PROCID \
+--rdzv_backend=c10d \
+--rdzv_endpoint="$MASTER_ADDR:$MASTER_PORT" \
+    fastvideo/training/ar_diffusion_pipeline.py \
+    "${parallel_args[@]}" \
+    "${model_args[@]}" \
+    "${dataset_args[@]}" \
+    "${training_args[@]}" \
+    "${optimizer_args[@]}" \
+    "${validation_args[@]}" \
+    "${miscellaneous_args[@]}"
@@ -0,0 +1,138 @@
+#!/bin/bash
+#SBATCH --job-name=wan_tf_ode_init_3333
+#SBATCH --nodes=8
+#SBATCH --ntasks=8
+#SBATCH --ntasks-per-node=1
+#SBATCH --gres=gpu:8
+#SBATCH --cpus-per-task=128
+#SBATCH --mem=1440G
+#SBATCH --output=wan_tf_ode_init_3333_output/wan_tf_ode_init_3333_%j.out
+#SBATCH --error=wan_tf_ode_init_3333_output/wan_tf_ode_init_3333_%j.err
+#SBATCH --exclusive
+
+# Basic Info
+export NCCL_P2P_DISABLE=1
+export TORCH_NCCL_ENABLE_MONITORING=0
+export NCCL_DEBUG_SUBSYS=INIT,NET
+# different cache dir for different processes
+# export TRITON_CACHE_DIR=/tmp/triton_cache_${SLURM_PROCID}
+export MASTER_PORT=29500
+export NODE_RANK=$SLURM_PROCID
+nodes=( $(scontrol show hostnames $SLURM_JOB_NODELIST) )
+export MASTER_ADDR=${nodes[0]}
+export TOKENIZERS_PARALLELISM=false
+# export WANDB_API_KEY="8d9f4b39abd68eb4e29f6fc010b7ee71a2207cde"
+export WANDB_API_KEY="50632ebd88ffd970521cec9ab4a1a2d7e85bfc45"
+# export WANDB_API_KEY='your_wandb_api_key_here'
+export WANDB_BASE_URL="https://api.wandb.ai"
+export WANDB_MODE=online
+export FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN
+
+export HF_HOME=/mnt/data/huggingface
+export CUDA_HOME=/usr/local/cuda
+export PATH="$CUDA_HOME/bin:$PATH"
+export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/home/shared-bin/local/usr/lib/aarch64-linux-gnu:$CUDA_HOME/lib64"
+export CUDACXX="$CUDA_HOME/bin/nvcc"
+
+source ~/miniconda3/bin/activate
+conda activate fastvideo
+
+# Configs
+NUM_GPUS=8
+NUM_TOTAL_GPUS=64
+
+MODEL_PATH="wlsaidhi/SFWan2.1-T2V-1.3B-Diffusers"
+DATA_DIR="/mnt/data/fv-tf-ode-preprocessing-6k-mixkit-wan-1.3b/"
+VALIDATION_DATASET_FILE="/mnt/weka/home/hao.zhang/wl/FastVideo/examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free/validation_64.json"
+# IP=[MASTER NODE IP]
+
+# Training arguments
+training_args=(
+  --tracker_project_name "wan_tf_ode_init"
+  --output_dir "/mnt/data/wan_tf_ode_init_3333_mixkit_shift5"
+  --wandb_run_name "wan_tf_ode_init_3333_mixkit_shift5"
+  --warp_denoising_step
+  --dmd_denoising_steps '1000,750,500,250'
+  --log_visualization
+  --visualization_steps 100
+  --max_train_steps 30000
+  --train_batch_size 1
+  --train_sp_batch_size 1
+  --gradient_accumulation_steps 1
+  --num_latent_t 21
+  --num_height 480
+  --num_width 832
+  --num_frames 81
+  --enable_gradient_checkpointing_type "full"
+  --init_weights_from_safetensors "/mnt/data/wan_ar_diffusion_3333_mixkit_shift5/checkpoint-2400/transformer/diffusion_pytorch_model.safetensors"
+)
+
+# Parallel arguments
+parallel_args=(
+  --num_gpus $NUM_TOTAL_GPUS # 64
+  --sp_size 1
+  --tp_size 1
+  --hsdp_replicate_dim 1 # 64
+  --hsdp_shard_dim $NUM_TOTAL_GPUS
+)
+
+# Model arguments
+model_args=(
+  --model_path $MODEL_PATH
+  --pretrained_model_name_or_path $MODEL_PATH
+  --text-encoder-cpu-offload
+)
+
+# Dataset arguments
+dataset_args=(
+  --data_path "$DATA_DIR" 
+  --dataloader_num_workers 4
+)
+
+# Optimizer arguments
+optimizer_args=(
+  --optimizer-type "adamw"
+  --learning_rate 1e-5
+  --mixed_precision "bf16"
+  --training_state_checkpointing_steps 500
+  --weight_decay 0.01
+  --max_grad_norm 1.0
+)
+
+validation_args=(
+  # --log_validation
+  --validation_dataset_file "$VALIDATION_DATASET_FILE"
+  --validation_steps 10
+  --validation_sampling_steps "4"
+  --validation_guidance_scale "1.0" # not used for dmd inference
+  # --vae_cpu_offload True
+)
+
+# Miscellaneous arguments
+miscellaneous_args=(
+  --inference_mode False
+  --checkpoints_total_limit 6
+  --training_cfg_rate 0.1
+  --multi_phased_distill_schedule "4000-1"
+  --not_apply_cfg_solver
+  --dit_precision "fp32"
+  --num_euler_timesteps 50
+  --ema_start_step 0
+  # --enable_gradient_checkpointing_type "full"
+)
+
+# If you do not have 32 GPUs and to fit in memory, you can: 1. increase sp_size. 2. reduce num_latent_t
+srun torchrun \
+--nnodes $SLURM_JOB_NUM_NODES \
+--nproc_per_node $NUM_GPUS \
+--node_rank $SLURM_PROCID \
+--rdzv_backend=c10d \
+--rdzv_endpoint="$MASTER_ADDR:$MASTER_PORT" \
+    fastvideo/training/ode_causal_pipeline.py \
+    "${parallel_args[@]}" \
+    "${model_args[@]}" \
+    "${dataset_args[@]}" \
+    "${training_args[@]}" \
+    "${optimizer_args[@]}" \
+    "${validation_args[@]}" \
+    "${miscellaneous_args[@]}"
@@ -1,23 +0,0 @@
-# LTX-2 Crush-Smol Example
-# TODO: Update this doc.
-
-These are e2e example scripts for finetuning LTX-2 on the crush-smol dataset.
-
-## Execute the following commands from `FastVideo/` to run training:
-
-### Download crush-smol dataset:
-
-`bash examples/training/finetune/ltx2/overfit/download_dataset.sh`
-
-### Preprocess the videos and captions into latents:
-
-`bash examples/training/finetune/ltx2/overfit/preprocess_ltx2_data_t2v_new.sh`
-
-### Edit the following file and run finetuning:
-
-`bash examples/training/finetune/ltx2/overfit/finetune_t2v.sh`
-
-Notes:
- Update `DATASET_PATH` in the preprocess script to point to your merged dataset root (`videos/` + `videos2caption.json`).
- `MODEL_PATH` should point to a local LTX-2 diffusers-style directory that contains `model_index.json` and `text_encoder/gemma`.
-
@@ -1,3 +0,0 @@
-# #!/bin/bash
-#
-python scripts/huggingface/download_hf.py --repo_id "wlsaidhi/crush-smol-merged" --local_dir "data/crush-smol" --repo_type "dataset"
@@ -1,95 +0,0 @@
-#!/bin/bash
-
-export WANDB_BASE_URL="https://api.wandb.ai"
-export WANDB_MODE=online
-export TOKENIZERS_PARALLELISM=false
-
-MODEL_PATH="Davids048/LTX2-Base-Diffusers"
-# Also can use simple 1 video for overfitting experiments.
-# DATA_DIR="/home/hal-jundas/codes/FastVideo/data/crush-smol"
-DATA_DIR="<PATH_TO_PROCESSED_DATASET>"
-VALIDATION_DATASET_FILE="$(dirname "$0")/validation.json"
-echo  VALIDATION_DATASET_FILE: $VALIDATION_DATASET_FILE
-NUM_GPUS=4
-OVERFIT_HEIGHT=480
-OVERFIT_WIDTH=832
-OVERFIT_FRAMES=73
-
-training_args=(
-  --tracker_project_name "ltx2_t2v_finetune"
-  --output_dir "checkpoints/ltx2_t2v_finetune"
-  --max_train_steps 5000
-  --train_batch_size 1
-  --train_sp_batch_size 1
-  --gradient_accumulation_steps 1
-  --num_latent_t 10
-  --num_height $OVERFIT_HEIGHT
-  --num_width $OVERFIT_WIDTH
-  --num_frames $OVERFIT_FRAMES
-  --ltx2-first-frame-conditioning-p 0.1
-  --enable_gradient_checkpointing_type "full"
-  --mode "finetuning"
-)
-
-parallel_args=(
-  --num_gpus $NUM_GPUS
-  --sp_size $NUM_GPUS
-  --tp_size 1
-  --hsdp_replicate_dim 1
-  --hsdp_shard_dim $NUM_GPUS
-)
-
-model_args=(
-  --model_path $MODEL_PATH
-  --pretrained_model_name_or_path $MODEL_PATH
-)
-
-dataset_args=(
-  --data_path $DATA_DIR
-  --dataloader_num_workers 1
-)
-
-validation_args=(
-  --log_validation
-  --validation_dataset_file $VALIDATION_DATASET_FILE
-  --validation_steps 50
-  --validation_sampling_steps "50"
-  --validation_guidance_scale "3.0"
-)
-
-optimizer_args=(
-  --learning_rate 1e-5
-  --mixed_precision "bf16"
-  --weight_only_checkpointing_steps 1000
-  --training_state_checkpointing_steps 1000
-  --weight_decay 1e-4
-  --max_grad_norm 1.0
-  --lr_scheduler "linear"
-)
-
-miscellaneous_args=(
-  --inference_mode False
-  --checkpoints_total_limit 3
-  --dit_precision "fp32"
-  --dit_cpu_offload False
-  --dit_layerwise_offload False
-  --text_encoder_cpu_offload False
-  --image_encoder_cpu_offload False
-  --vae_cpu_offload False
-)
-
-# NOTE: Setting this environment variable to TORCH_SDPA to avoid the issue of stacking that failed in flash attn. 
-export FASTVIDEO_ATTENTION_BACKEND=TORCH_SDPA
-
-
-torchrun \
-  --nnodes 1 \
-  --nproc_per_node $NUM_GPUS \
-    fastvideo/training/ltx2_training_pipeline.py \
-    "${parallel_args[@]}" \
-    "${model_args[@]}" \
-    "${dataset_args[@]}" \
-    "${training_args[@]}" \
-    "${optimizer_args[@]}" \
-    "${validation_args[@]}" \
-    "${miscellaneous_args[@]}"
@@ -1,80 +0,0 @@
-#!/bin/bash
-
-export WANDB_BASE_URL="https://api.wandb.ai"
-export WANDB_MODE=online
-export TOKENIZERS_PARALLELISM=false
-
-MODEL_PATH="/path/to/LTX-2"
-DATA_DIR="data/crush-smol"
-VALIDATION_DATASET_FILE="$(dirname "$0")/validation.json"
-NUM_GPUS=1
-
-training_args=(
-  --tracker_project_name "ltx2_t2v_lora_finetune"
-  --output_dir "checkpoints/ltx2_t2v_lora_finetune"
-  --max_train_steps 2000
-  --train_batch_size 1
-  --train_sp_batch_size 1
-  --gradient_accumulation_steps 8
-  --num_latent_t 10
-  --num_height 480
-  --num_width 832
-  --num_frames 77
-  --ltx2-first-frame-conditioning-p 0.1
-  --enable_gradient_checkpointing_type "full"
-)
-
-parallel_args=(
-  --num_gpus $NUM_GPUS
-  --sp_size $NUM_GPUS
-  --tp_size 1
-  --hsdp_replicate_dim 1
-  --hsdp_shard_dim $NUM_GPUS
-)
-
-model_args=(
-  --model_path $MODEL_PATH
-  --pretrained_model_name_or_path $MODEL_PATH
-)
-
-dataset_args=(
-  --data_path $DATA_DIR
-  --dataloader_num_workers 1
-)
-
-validation_args=(
-  --log_validation
-  --validation_dataset_file $VALIDATION_DATASET_FILE
-  --validation_steps 200
-  --validation_sampling_steps "50"
-  --validation_guidance_scale "3.0"
-)
-
-optimizer_args=(
-  --learning_rate 2e-4
-  --mixed_precision "bf16"
-  --weight_only_checkpointing_steps 1000
-  --training_state_checkpointing_steps 1000
-  --weight_decay 1e-4
-  --max_grad_norm 1.0
-  --lora_training True
-  --lora_rank 16
-  --lora_alpha 16
-)
-
-miscellaneous_args=(
-  --inference_mode False
-  --checkpoints_total_limit 3
-)
-
-torchrun \
-  --nnodes 1 \
-  --nproc_per_node $NUM_GPUS \
-    fastvideo/training/ltx2_training_pipeline.py \
-    "${parallel_args[@]}" \
-    "${model_args[@]}" \
-    "${dataset_args[@]}" \
-    "${training_args[@]}" \
-    "${optimizer_args[@]}" \
-    "${validation_args[@]}" \
-    "${miscellaneous_args[@]}"
@@ -1,35 +0,0 @@
-#!/bin/bash
-
-GPU_NUM=1
-MODEL_PATH="Davids048/LTX2-Base-Diffusers"
-# DATASET_PATH="data/overfit"
-DATASET_PATH="data/crush-smol"
-OUTPUT_DIR="$DATASET_PATH"
-WITH_AUDIO=true
-
-# Convert one-file overfit metadata into merged format if needed.
-if [ ! -f "$DATASET_PATH/videos2caption.json" ] && [ -f "$DATASET_PATH/overfit.json" ]; then
-  python scripts/dataset_preparation/convert_to_merged_dataset.py \
-    --items-json "$DATASET_PATH/overfit.json" \
-    --output-dir "$DATASET_PATH"
-fi
-
-
-torchrun --nproc_per_node=$GPU_NUM \
-    --master_port=29513 \
-    -m fastvideo.pipelines.preprocess.v1_preprocessing_new \
-    --model_path $MODEL_PATH \
-    --mode preprocess \
-    --workload_type t2v \
-    --preprocess.video_loader_type torchvision \
-    --preprocess.dataset_type merged \
-    --preprocess.dataset_path $DATASET_PATH \
-    --preprocess.dataset_output_dir $OUTPUT_DIR \
-    --preprocess.with_audio $WITH_AUDIO \
-    --preprocess.preprocess_video_batch_size 1 \
-    --preprocess.dataloader_num_workers 0 \
-    --preprocess.max_height 480 \
-    --preprocess.max_width 832 \
-    --preprocess.num_frames 73 \
-    --preprocess.train_fps 16 \
-    --preprocess.video_length_tolerance_range 5
@@ -1,13 +0,0 @@
-{
-  "data": [
-    {
-      "caption": "The camera opens in a calm, sunlit frog yoga studio. Warm morning light washes over the wooden floor as incense smoke drifts lazily in the air. The senior frog instructor sits cross-legged at the center, eyes closed, voice deep and calm. “We are one with the pond.” All the frogs answer softly: “Ommm...” “We are one with the mud.” “Ommm...” He smiles faintly. “We are one with the flies.” A quiet pause. The camera slowly pans to the side — one frog twitches, eyes darting. Suddenly — *thwip!* — its tongue snaps out, catching a fly mid-air and pulling it into its mouth. The master exhales slowly, still serene.",
-      "image_path": null,
-      "video_path": null,
-      "num_inference_steps": 50,
-      "height": 1088,
-      "width": 1920,
-      "num_frames": 121
-    }
-  ]
-}
@@ -1,31 +0,0 @@
-{
-  "data": [
-    {
-      "caption": "A large metal cylinder is seen pressing down on a pile of Oreo cookies, flattening them as if they were under a hydraulic press.",
-      "image_path": null,
-      "video_path": null,
-      "num_inference_steps": 50,
-      "height": 480,
-      "width": 832,
-      "num_frames": 77
-    },
-    {
-      "caption": "A large metal cylinder is seen compressing colorful clay into a compact shape, demonstrating the power of a hydraulic press.",
-      "image_path": null,
-      "video_path": null,
-      "num_inference_steps": 50,
-      "height": 480,
-      "width": 832,
-      "num_frames": 77
-    },
-    {
-      "caption": "A large metal cylinder is seen pressing down on a pile of colorful candies, flattening them as if they were under a hydraulic press. The candies are crushed and broken into small pieces, creating a mess on the table.",
-      "image_path": null,
-      "video_path": null,
-      "num_inference_steps": 50,
-      "height": 480,
-      "width": 832,
-      "num_frames": 77
-    }
-  ]
-}
@@ -0,0 +1 @@
+The raw mixkit dataset is downloaded and preprocessed following instructions from https://github.com/tianweiy/CausVid/tree/master
@@ -0,0 +1,25 @@
+#!/bin/bash
+
+GPU_NUM=1 # 2,4,8
+MODEL_PATH="Wan-AI/Wan2.1-T2V-1.3B-Diffusers"
+MODEL_TYPE="wan"
+DATA_MERGE_PATH="/mnt/data/mixkit/merge.txt"
+OUTPUT_DIR="/mnt/data/mixkit_wan_1.3b_processed_t2v/"
+
+torchrun --nproc_per_node=$GPU_NUM \
+    fastvideo/pipelines/preprocess/v1_preprocess.py \
+    --model_path $MODEL_PATH \
+    --data_merge_path $DATA_MERGE_PATH \
+    --preprocess_video_batch_size 1 \
+    --seed 42 \
+    --max_height 480 \
+    --max_width 832 \
+    --num_frames 81 \
+    --dataloader_num_workers 0 \
+    --output_dir=$OUTPUT_DIR \
+    --train_fps 16 \
+    --samples_per_file 8 \
+    --flush_frequency 8 \
+    --video_length_tolerance_range 5 \
+    --drop_short_ratio 0.0 \
+    --preprocess_task "t2v" 
@@ -0,0 +1,66 @@
+#!/bin/bash
+#SBATCH --nodes=1
+#SBATCH --ntasks=1
+#SBATCH --ntasks-per-node=1
+#SBATCH --gres=gpu:8
+#SBATCH --cpus-per-task=128
+#SBATCH --mem=1440G
+#SBATCH --output=preprocess_mixkit_wan_1.3b_t2v/preprocess_wan_data_t2v.log
+#SBATCH --error=preprocess_mixkit_wan_1.3b_t2v/preprocess_wan_data_t2v.error
+#SBATCH --exclusive
+
+# Basic Info
+export NCCL_P2P_DISABLE=1
+export TORCH_NCCL_ENABLE_MONITORING=0
+export NCCL_DEBUG_SUBSYS=INIT,NET
+# different cache dir for different processes
+# export TRITON_CACHE_DIR=/tmp/triton_cache_${SLURM_PROCID}
+export MASTER_PORT=29500
+export NODE_RANK=$SLURM_PROCID
+nodes=( $(scontrol show hostnames $SLURM_JOB_NODELIST) )
+export MASTER_ADDR=${nodes[0]}
+export TOKENIZERS_PARALLELISM=false
+# export WANDB_API_KEY="8d9f4b39abd68eb4e29f6fc010b7ee71a2207cde"
+export WANDB_API_KEY="50632ebd88ffd970521cec9ab4a1a2d7e85bfc45"
+# export WANDB_API_KEY='your_wandb_api_key_here'
+export WANDB_BASE_URL="https://api.wandb.ai"
+export WANDB_MODE=online
+export FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN
+
+export HF_HOME=/mnt/data/huggingface
+export CUDA_HOME=/usr/local/cuda
+export PATH="$CUDA_HOME/bin:$PATH"
+export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/home/shared-bin/local/usr/lib/aarch64-linux-gnu:$CUDA_HOME/lib64"
+export CUDACXX="$CUDA_HOME/bin/nvcc"
+
+source ~/miniconda3/bin/activate
+conda activate fastvideo
+
+GPU_NUM=1 # 2,4,8
+MODEL_PATH="Wan-AI/Wan2.1-T2V-1.3B-Diffusers"
+MODEL_TYPE="wan"
+DATA_MERGE_PATH="/mnt/data/mixkit/merge.txt"
+OUTPUT_DIR="/mnt/data/mixkit_wan_1.3b_processed_t2v/"
+
+srun torchrun \
+--nnodes $SLURM_JOB_NUM_NODES \
+--nproc_per_node $GPU_NUM \
+--node_rank $SLURM_PROCID \
+--rdzv_backend=c10d \
+--rdzv_endpoint="$MASTER_ADDR:$MASTER_PORT" \
+    fastvideo/pipelines/preprocess/v1_preprocess.py \
+    --model_path $MODEL_PATH \
+    --data_merge_path $DATA_MERGE_PATH \
+    --preprocess_video_batch_size 1 \
+    --seed 42 \
+    --max_height 480 \
+    --max_width 832 \
+    --num_frames 81 \
+    --dataloader_num_workers 1 \
+    --output_dir=$OUTPUT_DIR \
+    --train_fps 16 \
+    --samples_per_file 8 \
+    --flush_frequency 8 \
+    --video_length_tolerance_range 5 \
+    --drop_short_ratio 0.0 \
+    --preprocess_task "t2v" 
@@ -9,7 +9,7 @@ build-backend = "scikit_build_core.build"

 [project]
 name = "fastvideo-kernel"
-version = "0.2.6"
+version = "0.2.5"
 description = "Unified CUDA kernels for FastVideo"
 readme = "README.md"
 requires-python = ">=3.10"
@@ -287,8 +287,9 @@ def block_sparse_attn(
    block_sparse_fwd, block_sparse_bwd = _get_sm90_ops()
    if (not _force_triton()) and _is_sm90() and (block_sparse_fwd is not None) and (block_sparse_bwd is not None):
        return block_sparse_attn_sm90(q, k, v, block_map, variable_block_sizes)
-    # Triton path: supports q_seq_len != kv_seq_len as long as both are padded
-    # to a multiple of the block size (64 tokens).
+    # Triton path: generally assumes q/k/v share the same padded length
+    if q.shape[2] != k.shape[2] or q.shape[2] != v.shape[2]:
+        raise RuntimeError("Triton fallback requires q/k/v to have the same padded length.")
    return block_sparse_attn_triton(q, k, v, block_map, variable_block_sizes)


@@ -141,6 +141,12 @@ def video_sparse_attn(
        # Use autograd-enabled wrapper so backward works (and still uses SM90 kernel when available)
        out_s = block_sparse_attn(q, k, v, mask, variable_block_sizes)[0]
    else:
+        if q_seq_len != kv_seq_len:
+            raise RuntimeError(
+                "q/k have different lengths, but the compiled CUDA kernel (block_sparse_fwd) "
+                "is not available. The Triton fallback currently requires q and k/v to have "
+                "the same padded length."
+            )
        # Triton-only forward (kept for environments without the wrapper deps)
        out_s, _ = triton_block_sparse_attn_forward(q, k, v, idx, num, variable_block_sizes)

@@ -29,7 +29,7 @@ configs = [


 # ──────────────────────────── SPARSE ADDITION BEGIN ───────────────────────────
-@triton.autotune(configs, key=["N_CTX_Q", "HEAD_DIM"])
+@triton.autotune(configs, key=["N_CTX", "HEAD_DIM"])
@triton.jit
 def _attn_fwd_sparse(
        Q,
@@ -60,8 +60,7 @@ def _attn_fwd_sparse(
        stride_on,
        Z,
        H,
-        N_CTX_Q,  #
-        N_CTX_KV,  #
+        N_CTX,  #
        HEAD_DIM: tl.constexpr,  #
        BLOCK_M: tl.constexpr,
        BLOCK_N: tl.constexpr,
@@ -76,29 +75,24 @@ def _attn_fwd_sparse(
    off_hz = tl.program_id(1)  # fused (batch, head)
    b = off_hz // H
    h = off_hz % H
-    q_tiles = N_CTX_Q // BLOCK_M
+    q_tiles = N_CTX // BLOCK_M
    meta_base = ((b * H + h) * q_tiles + q_blk)

    kv_blocks = tl.load(q2k_num + meta_base)  # int32
    kv_ptr = q2k_index + meta_base * max_kv_blks  # ptr to list

    # ----- base pointers -----
-    # Note: when q and kv have different sequence lengths, their per-(batch,head)
-    # strides differ, so we must compute separate base offsets.
-    q_off = (b.to(tl.int64) * stride_qz + h.to(tl.int64) * stride_qh)
-    k_off = (b.to(tl.int64) * stride_kz + h.to(tl.int64) * stride_kh)
-    v_off = (b.to(tl.int64) * stride_vz + h.to(tl.int64) * stride_vh)
-    o_off = (b.to(tl.int64) * stride_oz + h.to(tl.int64) * stride_oh)
+    qvk_off = (b.to(tl.int64) * stride_qz + h.to(tl.int64) * stride_qh)

-    Q_ptr = tl.make_block_ptr(base=Q + q_off,
-                              shape=(N_CTX_Q, HEAD_DIM),
+    Q_ptr = tl.make_block_ptr(base=Q + qvk_off,
+                              shape=(N_CTX, HEAD_DIM),
                              strides=(stride_qm, stride_qk),
                              offsets=(q_blk * BLOCK_M, 0),
                              block_shape=(BLOCK_M, HEAD_DIM),
                              order=(1, 0))

-    K_base = tl.make_block_ptr(base=K + k_off,
-                               shape=(HEAD_DIM, N_CTX_KV),
+    K_base = tl.make_block_ptr(base=K + qvk_off,
+                               shape=(HEAD_DIM, N_CTX),
                               strides=(stride_kk, stride_kn),
                               offsets=(0, 0),
                               block_shape=(HEAD_DIM, BLOCK_N),
@@ -106,15 +100,15 @@ def _attn_fwd_sparse(

    v_order: tl.constexpr = (0, 1) if V.dtype.element_ty == tl.float8e5 else (1,
                                                                              0)
-    V_base = tl.make_block_ptr(base=V + v_off,
-                               shape=(N_CTX_KV, HEAD_DIM),
+    V_base = tl.make_block_ptr(base=V + qvk_off,
+                               shape=(N_CTX, HEAD_DIM),
                               strides=(stride_vk, stride_vn),
                               offsets=(0, 0),
                               block_shape=(BLOCK_N, HEAD_DIM),
                               order=v_order)

-    O_ptr = tl.make_block_ptr(base=Out + o_off,
-                              shape=(N_CTX_Q, HEAD_DIM),
+    O_ptr = tl.make_block_ptr(base=Out + qvk_off,
+                              shape=(N_CTX, HEAD_DIM),
                              strides=(stride_om, stride_on),
                              offsets=(q_blk * BLOCK_M, 0),
                              block_shape=(BLOCK_M, HEAD_DIM),
@@ -156,7 +150,7 @@ def _attn_fwd_sparse(
    # ----- epilogue -----
    m_i += tl.math.log2(l_i)
    acc = acc / l_i[:, None]
-    tl.store(M + off_hz * N_CTX_Q + offs_m, m_i)
+    tl.store(M + off_hz * N_CTX + offs_m, m_i)
    tl.store(O_ptr, acc.to(Out.type.element_ty))


@@ -207,7 +201,7 @@ def _attn_bwd_dkdv(
        stride_tok,
        stride_d,  #
        H,
-        N_CTX_KV,
+        N_CTX,
        BLOCK_M1: tl.constexpr,  #
        BLOCK_N1: tl.constexpr,  #
        HEAD_DIM: tl.constexpr,  #
@@ -227,8 +221,8 @@ def _attn_bwd_dkdv(
    off_hz = tl.program_id(2)  # fused (batch, head)
    b = off_hz // H
    h = off_hz % H
-    kv_tiles = N_CTX_KV // BLOCK_N1
-    meta_base = ((b * H + h) * kv_tiles + kv_blk)
+    q_tiles = N_CTX // BLOCK_N1
+    meta_base = ((b * H + h) * q_tiles + kv_blk)

    q_blocks = tl.load(k2q_num + meta_base)  # int32
    q_ptr = k2q_index + meta_base * max_q_blks  # ptr to list
@@ -308,21 +302,16 @@ def _attn_bwd_dq(

    kv_blocks = tl.load(q2k_num + meta_base)  # int32
    kv_ptr = q2k_index + meta_base * max_kv_blks  # ptr to list
+    block_size = tl.load(variable_block_sizes + q_blk)

    for blk_idx in range(kv_blocks * 2):
-        kv_idx = tl.load(kv_ptr + blk_idx // 2).to(tl.int32)
-        # variable_block_sizes is defined per KV block (tile). Mask must therefore
-        # use kv_idx (not q_blk). Also, because we split each 64-token block into
-        # two 32-token halves, the mask must account for the half-block offset.
-        block_size = tl.load(variable_block_sizes + kv_idx).to(tl.int32)
-        half = (blk_idx % 2).to(tl.int32)
-        block_sparse_offset = (kv_idx * 2 + half) * step_n * stride_tok
+        block_sparse_offset = (tl.load(kv_ptr + blk_idx // 2).to(tl.int32) * 2 +
+                               blk_idx % 2) * step_n * stride_tok
        kT = tl.load(kT_ptrs + block_sparse_offset)
        vT = tl.load(vT_ptrs + block_sparse_offset)
        qk = tl.dot(q, kT)
        p = tl.math.exp2(qk - m)
-        offs_in_block = half * step_n + tl.arange(0, BLOCK_N2)
-        mask = offs_in_block < block_size
+        mask = tl.arange(0, BLOCK_N2) < block_size.to(tl.int32)
        p = tl.where(mask[None, :], p, 0.0)
        # Compute dP and dS.
        dp = tl.dot(do, vT).to(tl.float32)
@@ -478,235 +467,19 @@ def _attn_bwd(
    tl.store(dq_ptrs, dq)


-@triton.jit
-def _attn_bwd_dkdv_kernel(
-        Q,
-        K,
-        V,
-        sm_scale,  #
-        DO,  #
-        DK,
-        DV,  #
-        M,
-        D,
-        k2q_index,
-        k2q_num,
-        max_q_blks,
-        variable_block_sizes,
-        # shared token/dim strides (assumed contiguous along token and dim)
-        stride_tok,
-        stride_d,  #
-        # batch/head strides (may differ between Q and KV)
-        stride_qz,
-        stride_qh,
-        stride_kz,
-        stride_kh,
-        stride_vz,
-        stride_vh,
-        stride_doz,
-        stride_doh,
-        stride_dkz,
-        stride_dkh,
-        stride_dvz,
-        stride_dvh,
-        H,
-        N_CTX_Q,
-        N_CTX_KV,
-        BLOCK_M1: tl.constexpr,  #
-        BLOCK_N1: tl.constexpr,  #
-        HEAD_DIM: tl.constexpr):
-    """
-    Backward kernel that computes dK and dV for each KV block (64 tokens).
-    Grid:
-      pid0: kv_blk in [0, N_CTX_KV/BLOCK_N1)
-      pid2: fused (batch, head) in [0, B*H)
-    """
-    bhid = tl.program_id(2)
-    b = bhid // H
-    h = bhid % H
-    kv_blk = tl.program_id(0)
-
-    q_adj = (b.to(tl.int64) * stride_qz + h.to(tl.int64) * stride_qh)
-    kv_adj_k = (b.to(tl.int64) * stride_kz + h.to(tl.int64) * stride_kh)
-    kv_adj_v = (b.to(tl.int64) * stride_vz + h.to(tl.int64) * stride_vh)
-    do_adj = (b.to(tl.int64) * stride_doz + h.to(tl.int64) * stride_doh)
-    dk_adj = (b.to(tl.int64) * stride_dkz + h.to(tl.int64) * stride_dkh)
-    dv_adj = (b.to(tl.int64) * stride_dvz + h.to(tl.int64) * stride_dvh)
-
-    Q = Q + q_adj
-    K = K + kv_adj_k
-    V = V + kv_adj_v
-    DO = DO + do_adj
-    DK = DK + dk_adj
-    DV = DV + dv_adj
-
-    # M and D (delta) are always sized by Q length.
-    M = M + (bhid * N_CTX_Q).to(tl.int64)
-    D = D + (bhid * N_CTX_Q).to(tl.int64)
-
-    offs_k = tl.arange(0, HEAD_DIM)
-    start_n = kv_blk * BLOCK_N1
-    offs_n = start_n + tl.arange(0, BLOCK_N1)
-
-    # load K and V: they stay in SRAM throughout the inner loop.
-    k = tl.load(K + offs_n[:, None] * stride_tok + offs_k[None, :] * stride_d)
-    v = tl.load(V + offs_n[:, None] * stride_tok + offs_k[None, :] * stride_d)
-
-    dv_acc = tl.zeros([BLOCK_N1, HEAD_DIM], dtype=tl.float32)
-    dk_acc = tl.zeros([BLOCK_N1, HEAD_DIM], dtype=tl.float32)
-
-    num_steps = N_CTX_Q // BLOCK_M1
-    dk_acc, dv_acc = _attn_bwd_dkdv(
-        dk_acc,
-        dv_acc,
-        Q,
-        k,
-        v,
-        sm_scale,
-        DO,
-        M,
-        D,
-        k2q_index,
-        k2q_num,
-        max_q_blks,
-        variable_block_sizes,
-        stride_tok,
-        stride_d,
-        H,
-        N_CTX_KV,
-        BLOCK_M1=BLOCK_M1,
-        BLOCK_N1=BLOCK_N1,
-        HEAD_DIM=HEAD_DIM,
-        start_n=start_n,
-        start_m=0,
-        num_steps=num_steps,
-    )
-
-    dv_ptrs = DV + offs_n[:, None] * stride_tok + offs_k[None, :] * stride_d
-    tl.store(dv_ptrs, dv_acc)
-
-    dk_acc *= sm_scale
-    dk_ptrs = DK + offs_n[:, None] * stride_tok + offs_k[None, :] * stride_d
-    tl.store(dk_ptrs, dk_acc)
-
-
-@triton.jit
-def _attn_bwd_dq_kernel(
-        Q,
-        K,
-        V,
-        DO,  #
-        DQ,
-        M,
-        D,
-        q2k_index,
-        q2k_num,
-        max_kv_blks,
-        variable_block_sizes,
-        # shared token/dim strides (assumed contiguous along token and dim)
-        stride_tok,
-        stride_d,  #
-        # batch/head strides (may differ between Q and KV)
-        stride_qz,
-        stride_qh,
-        stride_kz,
-        stride_kh,
-        stride_vz,
-        stride_vh,
-        stride_doz,
-        stride_doh,
-        stride_dqz,
-        stride_dqh,
-        H,
-        N_CTX_Q,
-        BLOCK_M2: tl.constexpr,  #
-        BLOCK_N2: tl.constexpr,  #
-        HEAD_DIM: tl.constexpr):
-    """
-    Backward kernel that computes dQ for each Q block (64 tokens).
-    Grid:
-      pid0: q_blk in [0, N_CTX_Q/BLOCK_M2)
-      pid2: fused (batch, head) in [0, B*H)
-    """
-    LN2 = 0.6931471824645996  # = ln(2)
-    bhid = tl.program_id(2)
-    b = bhid // H
-    h = bhid % H
-    q_blk = tl.program_id(0)
-
-    q_adj = (b.to(tl.int64) * stride_qz + h.to(tl.int64) * stride_qh)
-    kv_adj_k = (b.to(tl.int64) * stride_kz + h.to(tl.int64) * stride_kh)
-    kv_adj_v = (b.to(tl.int64) * stride_vz + h.to(tl.int64) * stride_vh)
-    do_adj = (b.to(tl.int64) * stride_doz + h.to(tl.int64) * stride_doh)
-    dq_adj = (b.to(tl.int64) * stride_dqz + h.to(tl.int64) * stride_dqh)
-
-    Q = Q + q_adj
-    K = K + kv_adj_k
-    V = V + kv_adj_v
-    DO = DO + do_adj
-    DQ = DQ + dq_adj
-
-    M = M + (bhid * N_CTX_Q).to(tl.int64)
-    D = D + (bhid * N_CTX_Q).to(tl.int64)
-
-    offs_k = tl.arange(0, HEAD_DIM)
-    start_m = q_blk * BLOCK_M2
-    offs_m = start_m + tl.arange(0, BLOCK_M2)
-
-    q = tl.load(Q + offs_m[:, None] * stride_tok + offs_k[None, :] * stride_d)
-    do = tl.load(DO + offs_m[:, None] * stride_tok + offs_k[None, :] * stride_d)
-    m = tl.load(M + offs_m)[:, None]
-
-    dq_acc = tl.zeros([BLOCK_M2, HEAD_DIM], dtype=tl.float32)
-    num_steps = 0  # unused in _attn_bwd_dq
-    dq_acc = _attn_bwd_dq(
-        dq_acc,
-        q,
-        K,
-        V,
-        do,
-        m,
-        D,
-        q2k_index,
-        q2k_num,
-        max_kv_blks,
-        variable_block_sizes,
-        stride_tok,
-        stride_d,
-        H,
-        N_CTX_Q,
-        BLOCK_M2=BLOCK_M2,
-        BLOCK_N2=BLOCK_N2,
-        HEAD_DIM=HEAD_DIM,
-        start_m=start_m,
-        start_n=0,
-        num_steps=num_steps,
-    )
-
-    dq_ptrs = DQ + offs_m[:, None] * stride_tok + offs_k[None, :] * stride_d
-    dq_acc *= LN2
-    tl.store(dq_ptrs, dq_acc)
-
-
 # ──────────────────────────── SPARSE ADDITION BEGIN ───────────────────────────
 def triton_block_sparse_attn_forward(q, k, v, q2k_index, q2k_num,
                                     variable_block_sizes):
-    B, H, Tq, D = q.shape
-    Tkv = k.shape[2]
+    B, H, T, D = q.shape
    sm_scale = 1.0 / math.sqrt(D)
    max_kv_blks = q2k_index.shape[-1]
-    assert Tq % 64 == 0, f"q length must be a multiple of 64, but got {Tq}"
-    assert Tkv % 64 == 0, f"kv length must be a multiple of 64, but got {Tkv}"
+    assert T % 64 == 0, f"T must be a multiple of 64, but got {T}"
    assert q2k_num.shape[
-        -1] == Tq // 64, f"shape mismatch, Tq // 64 = {Tq // 64}, q2k_num.shape[-2] = {q2k_num.shape[-2]}"
-    assert variable_block_sizes.numel() == Tkv // 64, (
-        f"shape mismatch, variable_block_sizes must have length {Tkv // 64}, "
-        f"got {variable_block_sizes.numel()}"
-    )
+        -1] == T // 64, f"shape mismatch, T // 64 = {T // 64}, q2k_num.shape[-2] = {q2k_num.shape[-2]}"
    o = torch.empty_like(q)
-    M = torch.empty((B, H, Tq), dtype=torch.float32, device=q.device)
+    M = torch.empty((B, H, T), dtype=torch.float32, device=q.device)

-    grid = lambda _: (triton.cdiv(Tq, 64), B * H, 1)
+    grid = lambda _: (triton.cdiv(T, 64), B * H, 1)
    _attn_fwd_sparse[grid](q,
                           k,
                           v,
@@ -735,8 +508,7 @@ def triton_block_sparse_attn_forward(q, k, v, q2k_index, q2k_num,
                           o.stride(3),
                           B,
                           H,
-                           Tq,
-                           Tkv,
+                           T,
                           HEAD_DIM=D,
                           STAGE=3)

@@ -746,21 +518,21 @@ def triton_block_sparse_attn_forward(q, k, v, q2k_index, q2k_num,
 def triton_block_sparse_attn_backward(do, q, k, v, o, M, q2k_index, q2k_num,
                                      k2q_index, k2q_num, variable_block_sizes):
    assert do.is_contiguous()
+    assert q.stride() == k.stride() == v.stride() == o.stride() == do.stride()

-    B, H, Tq, D = q.shape
-    Tkv = k.shape[2]
+    B, H, T, D = q.shape
    sm_scale = 1.0 / math.sqrt(D)
    dq = torch.empty_like(q)
    dk = torch.empty_like(k)
    dv = torch.empty_like(v)
-    BATCH, N_HEAD = q.shape[:2]
+    BATCH, N_HEAD, N_CTX = q.shape[:3]
    BLOCK_M1, BLOCK_N1, BLOCK_M2, BLOCK_N2 = 32, 64, 64, 32
    RCP_LN2 = 1.4426950408889634  # = 1.0 / ln(2)
    arg_k = k
    arg_k = arg_k * (sm_scale * RCP_LN2)
    PRE_BLOCK = 64
-    assert Tq % PRE_BLOCK == 0
-    pre_grid = (Tq // PRE_BLOCK, BATCH * N_HEAD)
+    assert N_CTX % PRE_BLOCK == 0
+    pre_grid = (N_CTX // PRE_BLOCK, BATCH * N_HEAD)
    delta = torch.empty_like(M)
    _attn_bwd_preprocess[pre_grid](
        o,
@@ -768,7 +540,7 @@ def triton_block_sparse_attn_backward(do, q, k, v, o, M, q2k_index, q2k_num,
        delta,  #
        BATCH,
        N_HEAD,
-        Tq,  #
+        N_CTX,  #
        BLOCK_M=PRE_BLOCK,
        HEAD_DIM=D  #
    )
@@ -776,75 +548,36 @@ def triton_block_sparse_attn_backward(do, q, k, v, o, M, q2k_index, q2k_num,
    max_q_blks = k2q_index.shape[-1]
    max_kv_blks = q2k_index.shape[-1]

-    # dK/dV kernel: grid over KV blocks
-    grid_kv = (Tkv // BLOCK_N1, 1, BATCH * N_HEAD)
-    _attn_bwd_dkdv_kernel[grid_kv](
+    grid = (N_CTX // BLOCK_N1, 1, BATCH * N_HEAD)
+    _attn_bwd[grid](
        q,
        arg_k,
        v,
        sm_scale,
        do,
+        dq,
        dk,
-        dv,
+        dv,  #
        M,
-        delta,
+        delta,  #
+        q2k_index,
+        q2k_num,
+        max_kv_blks,
        k2q_index,
        k2q_num,
        max_q_blks,
        variable_block_sizes,
-        q.stride(2),
-        q.stride(3),
        q.stride(0),
        q.stride(1),
-        arg_k.stride(0),
-        arg_k.stride(1),
-        v.stride(0),
-        v.stride(1),
-        do.stride(0),
-        do.stride(1),
-        dk.stride(0),
-        dk.stride(1),
-        dv.stride(0),
-        dv.stride(1),
+        q.stride(2),
+        q.stride(3),  #
        N_HEAD,
-        Tq,
-        Tkv,
+        N_CTX,  #
        BLOCK_M1=BLOCK_M1,
-        BLOCK_N1=BLOCK_N1,
-        HEAD_DIM=D,
-    )
-
-    # dQ kernel: grid over Q blocks
-    grid_q = (Tq // BLOCK_M2, 1, BATCH * N_HEAD)
-    _attn_bwd_dq_kernel[grid_q](
-        q,
-        arg_k,
-        v,
-        do,
-        dq,
-        M,
-        delta,
-        q2k_index,
-        q2k_num,
-        max_kv_blks,
-        variable_block_sizes,
-        q.stride(2),
-        q.stride(3),
-        q.stride(0),
-        q.stride(1),
-        arg_k.stride(0),
-        arg_k.stride(1),
-        v.stride(0),
-        v.stride(1),
-        do.stride(0),
-        do.stride(1),
-        dq.stride(0),
-        dq.stride(1),
-        N_HEAD,
-        Tq,
+        BLOCK_N1=BLOCK_N1,  #
        BLOCK_M2=BLOCK_M2,
-        BLOCK_N2=BLOCK_N2,
-        HEAD_DIM=D,
+        BLOCK_N2=BLOCK_N2,  #
+        HEAD_DIM=D  #
    )

    return dq, dk, dv
@@ -1 +1 @@
-__version__ = "0.2.6"
+__version__ = "0.2.5"
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: Apache-2.0

 import torch
-from sageattn3 import sageattn3_blackwell
+from fastvideo.attention.backends.sageattn.api import sageattn_blackwell

 from fastvideo.attention.backends.abstract import (AttentionBackend,
                                                   AttentionImpl,
@@ -67,6 +67,6 @@ class SageAttention3Impl(AttentionImpl):
        query = query.transpose(1, 2)
        key = key.transpose(1, 2)
        value = value.transpose(1, 2)
-        output = sageattn3_blackwell(query, key, value, is_causal=self.causal)
+        output = sageattn_blackwell(query, key, value, is_causal=self.causal)
        output = output.transpose(1, 2)
        return output
@@ -86,7 +86,6 @@ class PreprocessConfig:

    # Model configuration
    training_cfg_rate: float = 0.0
-    with_audio: bool = False

    # framework configuration
    seed: int = 42
@@ -191,10 +190,6 @@ class PreprocessConfig:
                                     type=float,
                                     default=PreprocessConfig.training_cfg_rate,
                                     help="Training CFG rate")
-        preprocess_args.add_argument(f"--{prefix_with_dot}with-audio",
-                                     action=StoreBoolean,
-                                     default=PreprocessConfig.with_audio,
-                                     help="Whether to extract and encode audio")
        preprocess_args.add_argument(f"--{prefix_with_dot}seed",
                                     type=int,
                                     default=PreprocessConfig.seed,
@@ -5,7 +5,6 @@ from fastvideo.configs.models.vaes.base import VAEConfig
 from fastvideo.configs.models.audio import (LTX2AudioDecoderConfig,
                                            LTX2AudioEncoderConfig,
                                            LTX2VocoderConfig)
-from fastvideo.configs.models.upsamplers.base import UpsamplerConfig

 __all__ = [
    "ModelConfig",
@@ -15,5 +14,4 @@ __all__ = [
    "LTX2AudioEncoderConfig",
    "LTX2AudioDecoderConfig",
    "LTX2VocoderConfig",
-    "UpsamplerConfig",
 ]
@@ -1,6 +1,5 @@
 from fastvideo.configs.models.dits.cosmos import CosmosVideoConfig
 from fastvideo.configs.models.dits.cosmos2_5 import Cosmos25VideoConfig
-from fastvideo.configs.models.dits.hunyuangamecraft import HunyuanGameCraftConfig
 from fastvideo.configs.models.dits.hunyuanvideo import HunyuanVideoConfig
 from fastvideo.configs.models.dits.hunyuanvideo15 import HunyuanVideo15Config
 from fastvideo.configs.models.dits.longcat import LongCatVideoConfig
@@ -10,8 +9,7 @@ from fastvideo.configs.models.dits.wanvideo import WanVideoConfig
 from fastvideo.configs.models.dits.hyworld import HYWorldConfig

 __all__ = [
-    "HunyuanVideoConfig", "HunyuanVideo15Config", "HunyuanGameCraftConfig",
-    "WanVideoConfig", "StepVideoConfig", "CosmosVideoConfig",
-    "Cosmos25VideoConfig", "LongCatVideoConfig", "LTX2VideoConfig",
-    "HYWorldConfig"
+    "HunyuanVideoConfig", "HunyuanVideo15Config", "WanVideoConfig",
+    "StepVideoConfig", "CosmosVideoConfig", "Cosmos25VideoConfig",
+    "LongCatVideoConfig", "LTX2VideoConfig", "HYWorldConfig"
 ]
@@ -1,163 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-"""
-Configuration for HunyuanGameCraft transformer model.
-
-HunyuanGameCraft extends HunyuanVideo with:
-1. CameraNet for camera/action conditioning
-2. 33 input channels (16 latent + 16 gt_latent + 1 mask)
-3. Mask-based conditioning for autoregressive generation
-"""
-from dataclasses import dataclass, field
-
-import torch
-
-from fastvideo.configs.models.dits.base import DiTArchConfig, DiTConfig
-
-
-def is_double_block(n: str, m) -> bool:
-    return "double" in n and str.isdigit(n.split(".")[-1])
-
-
-def is_single_block(n: str, m) -> bool:
-    return "single" in n and str.isdigit(n.split(".")[-1])
-
-
-def is_refiner_block(n: str, m) -> bool:
-    return "refiner" in n and str.isdigit(n.split(".")[-1])
-
-
-def is_txt_in(n: str, m) -> bool:
-    return n.split(".")[-1] == "txt_in"
-
-
-def is_camera_net(n: str, m) -> bool:
-    return "camera_net" in n
-
-
-@dataclass
-class HunyuanGameCraftArchConfig(DiTArchConfig):
-    """Architecture config for HunyuanGameCraft transformer."""
-
-    # Version field for compatibility with saved config.json
-    _fastvideo_version: str = "0.1.0"
-
-    # Camera net flag (for config.json compatibility)
-    camera_net: bool = True
-
-    _fsdp_shard_conditions: list = field(
-        default_factory=lambda:
-        [is_double_block, is_single_block, is_refiner_block, is_camera_net])
-
-    _compile_conditions: list = field(
-        default_factory=lambda: [is_double_block, is_single_block, is_txt_in])
-
-    # Parameter names mapping from official checkpoint to FastVideo naming
-    # GameCraft weights are already close to FastVideo format with minor adjustments
-    param_names_mapping: dict = field(
-        default_factory=lambda: {
-            # MLP naming: fc1 -> fc_in, fc2 -> fc_out
-            r"^(.*)\.img_mlp\.fc1\.(.*)$":
-            r"\1.img_mlp.fc_in.\2",
-            r"^(.*)\.img_mlp\.fc2\.(.*)$":
-            r"\1.img_mlp.fc_out.\2",
-            r"^(.*)\.txt_mlp\.fc1\.(.*)$":
-            r"\1.txt_mlp.fc_in.\2",
-            r"^(.*)\.txt_mlp\.fc2\.(.*)$":
-            r"\1.txt_mlp.fc_out.\2",
-
-            # Single block MLP naming
-            r"^single_blocks\.(\d+)\.mlp\.fc1\.(.*)$":
-            r"single_blocks.\1.mlp.fc_in.\2",
-            r"^single_blocks\.(\d+)\.mlp\.fc2\.(.*)$":
-            r"single_blocks.\1.mlp.fc_out.\2",
-
-            # Token refiner naming
-            r"^txt_in\.individual_token_refiner\.blocks\.(\d+)\.(.*)$":
-            r"txt_in.refiner_blocks.\1.\2",
-
-            # Vector in naming
-            r"^vector_in\.in_layer\.(.*)$":
-            r"vector_in.fc_in.\1",
-            r"^vector_in\.out_layer\.(.*)$":
-            r"vector_in.fc_out.\1",
-
-            # Time embedder naming
-            r"^time_in\.mlp\.0\.(.*)$":
-            r"time_in.mlp.fc_in.\1",
-            r"^time_in\.mlp\.2\.(.*)$":
-            r"time_in.mlp.fc_out.\1",
-
-            # Guidance embedder naming (if present)
-            r"^guidance_in\.mlp\.0\.(.*)$":
-            r"guidance_in.mlp.fc_in.\1",
-            r"^guidance_in\.mlp\.2\.(.*)$":
-            r"guidance_in.mlp.fc_out.\1",
-
-            # Final layer adaLN modulation
-            r"^final_layer\.adaLN_modulation\.1\.(.*)$":
-            r"final_layer.adaLN_modulation.linear.\1",
-
-            # Refiner block MLP naming
-            r"^txt_in\.refiner_blocks\.(\d+)\.mlp\.fc1\.(.*)$":
-            r"txt_in.refiner_blocks.\1.mlp.fc_in.\2",
-            r"^txt_in\.refiner_blocks\.(\d+)\.mlp\.fc2\.(.*)$":
-            r"txt_in.refiner_blocks.\1.mlp.fc_out.\2",
-
-            # Camera net weights are already correctly named
-        })
-
-    # Reverse mapping for saving checkpoints
-    reverse_param_names_mapping: dict = field(default_factory=lambda: {})
-
-    # Model architecture parameters
-    # patch_size can be int or tuple - if tuple, it's [T, H, W]
-    patch_size: int | tuple[int, int, int] = 2
-    patch_size_t: int = 1
-    in_channels: int = 33  # 16 latent + 16 gt_latent + 1 mask
-    out_channels: int = 16
-    num_attention_heads: int = 24
-    attention_head_dim: int = 128
-    mlp_ratio: float = 4.0
-    num_layers: int = 20  # Double stream blocks
-    num_single_layers: int = 40  # Single stream blocks
-    num_refiner_layers: int = 2
-    rope_axes_dim: tuple[int, int, int] = (16, 56, 56)
-    guidance_embeds: bool = False  # GameCraft doesn't use guidance
-    dtype: torch.dtype | None = None
-    text_embed_dim: int = 4096  # LLaMA-3 hidden size
-    pooled_projection_dim: int = 768  # CLIP pooled output dim
-    rope_theta: int = 256
-    qk_norm: str = "rms_norm"
-
-    # Camera net parameters
-    camera_in_channels: int = 6  # Plücker coordinates
-    camera_downscale_coef: int = 8
-    camera_out_channels: int = 16
-
-    # Layers to exclude from LoRA
-    exclude_lora_layers: list[str] = field(
-        default_factory=lambda:
-        ["img_in", "txt_in", "time_in", "vector_in", "camera_net"])
-
-    def __post_init__(self):
-        super().__post_init__()
-        self.hidden_size: int = self.attention_head_dim * self.num_attention_heads
-        self.num_channels_latents: int = 16  # Output is 16 channels
-
-        # Convert patch_size list to tuple if needed (from JSON deserialization)
-        if isinstance(self.patch_size, list):
-            self.patch_size = tuple(self.patch_size)
-
-        # Convert rope_axes_dim list to tuple if needed
-        if isinstance(self.rope_axes_dim, list):
-            self.rope_axes_dim = tuple(self.rope_axes_dim)
-
-
-@dataclass
-class HunyuanGameCraftConfig(DiTConfig):
-    """Full config for HunyuanGameCraft model."""
-
-    arch_config: DiTArchConfig = field(
-        default_factory=HunyuanGameCraftArchConfig)
-
-    prefix: str = "HunyuanGameCraft"
@@ -26,6 +26,9 @@ class HunyuanVideo15ArchConfig(DiTArchConfig):

    param_names_mapping: dict = field(
        default_factory=lambda: {
+            r"^cond_type_embed\.(.*)$":
+            r"cond_type_embed.\1",
+
            # 1. context_embedder.time_text_embed submodules (specific rules, applied first):
            r"^context_embedder\.time_text_embed\.timestep_embedder\.linear_1\.(.*)$":
            r"txt_in.t_embedder.mlp.fc_in.\1",
@@ -57,6 +60,14 @@ class HunyuanVideo15ArchConfig(DiTArchConfig):
            r"txt_in.refiner_blocks.\1.adaLN_modulation.linear.\2",
            r"^context_embedder\.token_refiner\.refiner_blocks\.(\d+)\.self_attn_qkv\.(.*)$":
            r"txt_in.refiner_blocks.\1.self_attn_qkv.\2",
+            r"^image_embedder\.linear_1\.(.*)$":
+            r"image_embedder.linear_1.\1",
+            r"^image_embedder\.linear_2\.(.*)$":
+            r"image_embedder.linear_2.\1",
+            r"^image_embedder\.norm_in\.(.*)$":
+            r"image_embedder.norm_in.\1",
+            r"^image_embedder\.norm_out\.(.*)$":
+            r"image_embedder.norm_out.\1",

            # 2. txt_in_2 mapping:
            r"^context_embedder_2\.(.*)$":
@@ -146,6 +157,12 @@ class HunyuanVideo15ArchConfig(DiTArchConfig):
    exclude_lora_layers: list[str] = field(
        default_factory=lambda: ["img_in", "txt_in", "time_in", "vector_in"])

+    # Causal HunyuanVideo1.5
+    local_attn_size: int = -1  # Window size for temporal local attention (-1 indicates global attention)
+    sink_size: int = 0  # Size of the attention sink, we keep the first `sink_size` frames unchanged when rolling the KV cache
+    num_frames_per_block: int = 3
+    sliding_window_num_frames: int = 31
+
    def __post_init__(self):
        super().__post_init__()
        self.hidden_size: int = self.attention_head_dim * self.num_attention_heads
@@ -8,8 +8,12 @@ def is_double_block(n: str, m) -> bool:
    return "double" in n and str.isdigit(n.split(".")[-1])


-def is_refiner_block(n: str, m) -> bool:
-    return "refiner" in n and str.isdigit(n.split(".")[-1])
+def is_single_block(n: str, m) -> bool:
+    return "single" in n and str.isdigit(n.split(".")[-1])
+
+
+# def is_refiner_block(n: str, m) -> bool:
+#     return "refiner" in n and str.isdigit(n.split(".")[-1])


 def is_txt_in(n: str, m) -> bool:
@@ -19,10 +23,10 @@ def is_txt_in(n: str, m) -> bool:
@dataclass
 class HYWorldArchConfig(DiTArchConfig):
    _fsdp_shard_conditions: list = field(
-        default_factory=lambda: [is_double_block, is_refiner_block])
+        default_factory=lambda: [is_double_block, is_single_block])

    _compile_conditions: list = field(
-        default_factory=lambda: [is_double_block, is_refiner_block, is_txt_in])
+        default_factory=lambda: [is_double_block, is_single_block, is_txt_in])

    param_names_mapping: dict = field(
        default_factory=lambda: {
@@ -50,7 +54,8 @@ class HYWorldArchConfig(DiTArchConfig):
            r"^txt_in\.individual_token_refiner\.blocks\.(\d+)\.adaLN_modulation\.1\.(.*)$":
            r"txt_in.refiner_blocks.\1.adaLN_modulation.linear.\2",

-            # 2. time_in mappings:
+            # 2. time_in mappings (HYWorld uses TimestepEmbedder directly,
+            #    but FastVideo model inherits HunyuanVideo15TimeEmbedding with timestep_embedder):
            r"^time_in\.mlp\.0\.(.*)$":
            r"time_in.timestep_embedder.mlp.fc_in.\1",
            r"^time_in\.mlp\.2\.(.*)$":
@@ -1,110 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-from dataclasses import dataclass, field
-
-from fastvideo.configs.models.dits.base import DiTArchConfig, DiTConfig
-
-
-def is_blocks(n: str, m) -> bool:
-    return "blocks" in n and str.isdigit(n.split(".")[-1])
-
-
-@dataclass
-class LingBotWorldArchConfig(DiTArchConfig):
-    _fsdp_shard_conditions: list = field(default_factory=lambda: [is_blocks])
-
-    param_names_mapping: dict = field(
-        default_factory=lambda: {
-            r"^patch_embedding\.(.*)$": r"patch_embedding.proj.\1",
-            r"^patch_embedding_wancamctrl\.(.*)$":
-            r"patch_embedding_wancamctrl.proj.\1",
-            r"^c2ws_hidden_states_layer1\.(.*)$": r"c2ws_mlp.fc_in.\1",
-            r"^c2ws_hidden_states_layer2\.(.*)$": r"c2ws_mlp.fc_out.\1",
-            r"^text_embedding\.0\.(.*)$":
-            r"condition_embedder.text_embedder.fc_in.\1",
-            r"^text_embedding\.2\.(.*)$":
-            r"condition_embedder.text_embedder.fc_out.\1",
-            r"^time_embedding\.0\.(.*)$":
-            r"condition_embedder.time_embedder.mlp.fc_in.\1",
-            r"^time_embedding\.2\.(.*)$":
-            r"condition_embedder.time_embedder.mlp.fc_out.\1",
-            r"^time_projection\.1\.(.*)$":
-            r"condition_embedder.time_modulation.linear.\1",
-            r"^blocks\.(\d+)\.modulation$": r"blocks.\1.scale_shift_table",
-            r"^blocks\.(\d+)\.self_attn\.q\.(.*)$": r"blocks.\1.to_q.\2",
-            r"^blocks\.(\d+)\.self_attn\.k\.(.*)$": r"blocks.\1.to_k.\2",
-            r"^blocks\.(\d+)\.self_attn\.v\.(.*)$": r"blocks.\1.to_v.\2",
-            r"^blocks\.(\d+)\.self_attn\.o\.(.*)$": r"blocks.\1.to_out.\2",
-            r"^blocks\.(\d+)\.self_attn\.norm_q\.(.*)$": r"blocks.\1.norm_q.\2",
-            r"^blocks\.(\d+)\.self_attn\.norm_k\.(.*)$": r"blocks.\1.norm_k.\2",
-            r"^blocks\.(\d+)\.norm3\.(.*)$":
-            r"blocks.\1.self_attn_residual_norm.norm.\2",
-            r"^blocks\.(\d+)\.cross_attn\.q\.(.*)$": r"blocks.\1.attn2.to_q.\2",
-            r"^blocks\.(\d+)\.cross_attn\.k\.(.*)$": r"blocks.\1.attn2.to_k.\2",
-            r"^blocks\.(\d+)\.cross_attn\.v\.(.*)$": r"blocks.\1.attn2.to_v.\2",
-            r"^blocks\.(\d+)\.cross_attn\.o\.(.*)$":
-            r"blocks.\1.attn2.to_out.\2",
-            r"^blocks\.(\d+)\.cross_attn\.norm_q\.(.*)$":
-            r"blocks.\1.attn2.norm_q.\2",
-            r"^blocks\.(\d+)\.cross_attn\.norm_k\.(.*)$":
-            r"blocks.\1.attn2.norm_k.\2",
-            r"^blocks\.(\d+)\.ffn\.0\.(.*)$": r"blocks.\1.ffn.fc_in.\2",
-            r"^blocks\.(\d+)\.ffn\.2\.(.*)$": r"blocks.\1.ffn.fc_out.\2",
-            r"^blocks\.(\d+)\.cam_injector_layer1\.(.*)$":
-            r"blocks.\1.cam_conditioner.cam_injector.fc_in.\2",
-            r"^blocks\.(\d+)\.cam_injector_layer2\.(.*)$":
-            r"blocks.\1.cam_conditioner.cam_injector.fc_out.\2",
-            r"^blocks\.(\d+)\.cam_scale_layer\.(.*)$":
-            r"blocks.\1.cam_conditioner.cam_scale_layer.\2",
-            r"^blocks\.(\d+)\.cam_shift_layer\.(.*)$":
-            r"blocks.\1.cam_conditioner.cam_shift_layer.\2",
-            r"^head\.modulation$": r"scale_shift_table",
-            r"^head\.head\.(.*)$": r"proj_out.\1",
-        })
-
-    # Reverse mapping for saving checkpoints: custom -> hf
-    reverse_param_names_mapping: dict = field(default_factory=lambda: {})
-
-    # Some LoRA adapters use the original official layer names instead of hf layer names,
-    # so apply this before the param_names_mapping
-    lora_param_names_mapping: dict = field(default_factory=lambda: {})
-
-    patch_size: tuple[int, int, int] = (1, 2, 2)
-    text_len: int = 512
-    num_attention_heads: int = 40
-    attention_head_dim: int = 128
-    in_channels: int = 16
-    out_channels: int = 16
-    text_dim: int = 4096
-    freq_dim: int = 256
-    ffn_dim: int = 13824
-    num_layers: int = 40
-    cross_attn_norm: bool = True
-    qk_norm: str = "rms_norm_across_heads"
-    eps: float = 1e-6
-    image_dim: int | None = None
-    added_kv_proj_dim: int | None = None
-    rope_max_seq_len: int = 1024
-    pos_embed_seq_len: int | None = None
-    exclude_lora_layers: list[str] = field(default_factory=lambda: ["embedder"])
-
-    # Wan MoE
-    boundary_ratio: float | None = None
-
-    # Causal Wan
-    local_attn_size: int = -1  # Window size for temporal local attention (-1 indicates global attention)
-    sink_size: int = 0  # Size of the attention sink, we keep the first `sink_size` frames unchanged when rolling the KV cache
-    num_frames_per_block: int = 3
-    sliding_window_num_frames: int = 21
-
-    def __post_init__(self):
-        super().__post_init__()
-        self.out_channels = self.out_channels or self.in_channels
-        self.hidden_size = self.num_attention_heads * self.attention_head_dim
-        self.num_channels_latents = self.out_channels
-
-
-@dataclass
-class LingBotWorldVideoConfig(DiTConfig):
-    arch_config: DiTArchConfig = field(default_factory=LingBotWorldArchConfig)
-
-    prefix: str = "Wan"
@@ -7,12 +7,10 @@ from dataclasses import dataclass, field

 from fastvideo.configs.models.dits.base import DiTArchConfig, DiTConfig

-import re
-

 def is_ltx2_blocks(name: str, _module) -> bool:
-    res = re.search(r"(?:^|\.)transformer_blocks\.\d+$", name) is not None
-    return res
+    """FSDP shard condition for LTX-2 transformer blocks."""
+    return "transformer_blocks" in name


@dataclass
@@ -1,6 +1,4 @@
 from dataclasses import dataclass, field
-
-import torch
 from fastvideo.configs.models.dits.wanvideo import WanVideoArchConfig, WanVideoConfig


@@ -78,14 +76,8 @@ class MatrixGameWanVideoArchConfig(WanVideoArchConfig):
    image_dim: int = 1280


-def _is_transformer_block(param_name: str, module: torch.nn.Module) -> bool:
-    return bool("blocks" in param_name and param_name.split(".")[-1].isdigit())
-
-
@dataclass
 class MatrixGameWanVideoConfig(WanVideoConfig):
    arch_config: MatrixGameWanVideoArchConfig = field(
        default_factory=MatrixGameWanVideoArchConfig)
    prefix: str = "Wan"
-    _compile_conditions: list = field(
-        default_factory=lambda: [_is_transformer_block])
@@ -1,31 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-
-from dataclasses import dataclass, field
-
-from fastvideo.configs.models.dits.base import DiTArchConfig, DiTConfig
-
-
-@dataclass
-class SD3Transformer2DArchConfig(DiTArchConfig):
-    # Diffusers SD3Transformer2DModel config fields.
-    sample_size: int = 128
-    patch_size: int = 2
-    num_layers: int = 24
-    attention_head_dim: int = 64
-    joint_attention_dim: int = 4096
-    caption_projection_dim: int = 1536
-    pooled_projection_dim: int = 2048
-    pos_embed_max_size: int = 384
-    dual_attention_layers: list[int] = field(
-        default_factory=lambda: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12])
-    qk_norm: str = "rms_norm"
-    in_channels: int = 16
-    out_channels: int = 16
-    num_attention_heads = 24
-
-
-@dataclass
-class SD3DiTConfig(DiTConfig):
-    arch_config: DiTArchConfig = field(
-        default_factory=SD3Transformer2DArchConfig)
-    prefix: str = "sd3"
@@ -1,6 +0,0 @@
-from fastvideo.configs.models.upsamplers.hunyuan15 import SRTo720pUpsamplerConfig, SRTo1080pUpsamplerConfig
-from fastvideo.configs.models.upsamplers.base import UpsamplerConfig
-
-__all__ = [
-    "SRTo720pUpsamplerConfig", "SRTo1080pUpsamplerConfig", "UpsamplerConfig"
-]
@@ -1,7 +0,0 @@
-from dataclasses import dataclass
-from fastvideo.configs.models.base import ModelConfig
-
-
-@dataclass
-class UpsamplerConfig(ModelConfig):
-    pass
@@ -1,20 +0,0 @@
-from dataclasses import dataclass
-from fastvideo.configs.models.upsamplers.base import UpsamplerConfig
-
-
-@dataclass
-class SRTo720pUpsamplerConfig(UpsamplerConfig):
-    in_channels: int = 0
-    out_channels: int = 0
-    hidden_channels: int = 64
-    num_blocks: int = 6
-    global_residual: bool = False
-
-
-@dataclass
-class SRTo1080pUpsamplerConfig(UpsamplerConfig):
-    z_channels: int = 0
-    out_channels: int = 0
-    block_out_channels: tuple[int, ...] = (0, 0)
-    num_res_blocks: int = 2
-    is_residual: bool = False
@@ -1,6 +1,5 @@
 from fastvideo.configs.models.vaes.cosmosvae import CosmosVAEConfig
 from fastvideo.configs.models.vaes.cosmos2_5vae import Cosmos25VAEConfig
-from fastvideo.configs.models.vaes.gamecraftvae import GameCraftVAEConfig
 from fastvideo.configs.models.vaes.hunyuanvae import HunyuanVAEConfig
 from fastvideo.configs.models.vaes.hunyuan15vae import Hunyuan15VAEConfig
 from fastvideo.configs.models.vaes.ltx2vae import LTX2VAEConfig
@@ -8,7 +7,6 @@ from fastvideo.configs.models.vaes.stepvideovae import StepVideoVAEConfig
 from fastvideo.configs.models.vaes.wanvae import WanVAEConfig

 __all__ = [
-    "GameCraftVAEConfig",
    "HunyuanVAEConfig",
    "WanVAEConfig",
    "StepVideoVAEConfig",
@@ -1,39 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-
-from dataclasses import dataclass, field
-
-import torch
-
-from fastvideo.configs.models.vaes.base import VAEArchConfig, VAEConfig
-
-
-@dataclass
-class AutoencoderKLArchConfig(VAEArchConfig):
-    _name_or_path: str = ""
-    act_fn: str = "silu"
-    block_out_channels: tuple[int, ...] | list[int] = field(
-        default_factory=list)
-    down_block_types: tuple[str, ...] | list[str] = field(default_factory=list)
-    up_block_types: tuple[str, ...] | list[str] = field(default_factory=list)
-    force_upcast: bool = True
-    in_channels: int = 3
-    latent_channels: int = 4
-    latents_mean: tuple[float, ...] | list[float] | None = None
-    latents_std: tuple[float, ...] | list[float] | None = None
-    layers_per_block: int = 1
-    mid_block_add_attention: bool = True
-    norm_num_groups: int = 32
-    out_channels: int = 3
-    sample_size: int = 32
-    scaling_factor: float | torch.Tensor = 0.18215
-    shift_factor: float | None = None
-    use_post_quant_conv: bool = True
-    use_quant_conv: bool = True
-
-    temporal_compression_ratio: int = 1
-    spatial_compression_ratio: int = 8
-
-
-@dataclass
-class AutoencoderKLVAEConfig(VAEConfig):
-    arch_config: VAEArchConfig = field(default_factory=AutoencoderKLArchConfig)
@@ -1,50 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-"""
-GameCraft VAE config - matches official config.json from Hunyuan-GameCraft-1.0.
-"""
-from dataclasses import dataclass, field
-
-from fastvideo.configs.models.vaes.base import VAEArchConfig, VAEConfig
-
-
-@dataclass
-class GameCraftVAEArchConfig(VAEArchConfig):
-    """Architecture config matching official AutoencoderKLCausal3D config.json."""
-
-    in_channels: int = 3
-    out_channels: int = 3
-    latent_channels: int = 16
-    down_block_types: tuple[str, ...] = (
-        "DownEncoderBlockCausal3D",
-        "DownEncoderBlockCausal3D",
-        "DownEncoderBlockCausal3D",
-        "DownEncoderBlockCausal3D",
-    )
-    up_block_types: tuple[str, ...] = (
-        "UpDecoderBlockCausal3D",
-        "UpDecoderBlockCausal3D",
-        "UpDecoderBlockCausal3D",
-        "UpDecoderBlockCausal3D",
-    )
-    block_out_channels: tuple[int, ...] = (128, 256, 512, 512)
-    layers_per_block: int = 2
-    act_fn: str = "silu"
-    norm_num_groups: int = 32
-    scaling_factor: float = 0.476986
-    spatial_compression_ratio: int = 8
-    temporal_compression_ratio: int = 4
-    time_compression_ratio: int = 4  # alias for DecoderCausal3D
-    mid_block_add_attention: bool = True
-    mid_block_causal_attn: bool = True
-    sample_size: int = 256  # from config.json
-    sample_tsize: int = 64  # from config.json
-
-    def __post_init__(self):
-        self.spatial_compression_ratio = 2**(len(self.block_out_channels) - 1)
-
-
-@dataclass
-class GameCraftVAEConfig(VAEConfig):
-    """Full config for GameCraft VAE."""
-
-    arch_config: VAEArchConfig = field(default_factory=GameCraftVAEArchConfig)
@@ -4,20 +4,20 @@ from fastvideo.configs.pipelines.cosmos import CosmosConfig
 from fastvideo.configs.pipelines.cosmos2_5 import Cosmos25Config
 from fastvideo.configs.pipelines.hunyuan import FastHunyuanConfig, HunyuanConfig
 from fastvideo.configs.pipelines.hunyuan15 import Hunyuan15T2V480PConfig, Hunyuan15T2V720PConfig
-from fastvideo.configs.pipelines.hunyuangamecraft import HunyuanGameCraftPipelineConfig
 from fastvideo.configs.pipelines.hyworld import HYWorldConfig
 from fastvideo.configs.pipelines.ltx2 import LTX2T2VConfig
-from fastvideo.registry import get_pipeline_config_cls_from_name
+from fastvideo.configs.pipelines.registry import (
+    get_pipeline_config_cls_from_name)
 from fastvideo.configs.pipelines.stepvideo import StepVideoT2VConfig
 from fastvideo.configs.pipelines.wan import (SelfForcingWanT2V480PConfig,
                                             WanI2V480PConfig, WanI2V720PConfig,
                                             WanT2V480PConfig, WanT2V720PConfig)

 __all__ = [
-    "HunyuanConfig", "FastHunyuanConfig", "HunyuanGameCraftPipelineConfig",
-    "PipelineConfig", "Hunyuan15T2V480PConfig", "Hunyuan15T2V720PConfig",
-    "SlidingTileAttnConfig", "WanT2V480PConfig", "WanI2V480PConfig",
-    "WanT2V720PConfig", "WanI2V720PConfig", "StepVideoT2VConfig",
-    "SelfForcingWanT2V480PConfig", "CosmosConfig", "Cosmos25Config",
-    "LTX2T2VConfig", "HYWorldConfig", "get_pipeline_config_cls_from_name"
+    "HunyuanConfig", "FastHunyuanConfig", "PipelineConfig",
+    "Hunyuan15T2V480PConfig", "Hunyuan15T2V720PConfig", "SlidingTileAttnConfig",
+    "WanT2V480PConfig", "WanI2V480PConfig", "WanT2V720PConfig",
+    "WanI2V720PConfig", "StepVideoT2VConfig", "SelfForcingWanT2V480PConfig",
+    "CosmosConfig", "Cosmos25Config", "LTX2T2VConfig", "HYWorldConfig",
+    "get_pipeline_config_cls_from_name"
 ]
@@ -8,7 +8,7 @@ from typing import Any, cast
 import torch

 from fastvideo.configs.models import (DiTConfig, EncoderConfig, ModelConfig,
-                                      VAEConfig, UpsamplerConfig)
+                                      VAEConfig)
 from fastvideo.configs.models.encoders import BaseEncoderOutput
 from fastvideo.configs.utils import update_config_from_args
 from fastvideo.logger import init_logger
@@ -44,15 +44,12 @@ class PipelineConfig:
    # Video generation parameters
    embedded_cfg_scale: float = 6.0
    flow_shift: float | None = None
-    flow_shift_sr: float | None = None
    disable_autocast: bool = False
    is_causal: bool = False

    # Model configuration
    dit_config: DiTConfig = field(default_factory=DiTConfig)
    dit_precision: str = "bf16"
-    upsampler_config: UpsamplerConfig = field(default_factory=UpsamplerConfig)
-    upsampler_precision: str = "fp32"

    # VAE configuration
    vae_config: VAEConfig = field(default_factory=VAEConfig)
@@ -219,24 +216,6 @@ class PipelineConfig:
            "Comma-separated list of denoising steps (e.g., '1000,757,522')",
        )

-        # STA (Sliding Tile Attention) parameters
-        parser.add_argument(
-            f"--{prefix_with_dot}STA-mode",
-            type=str,
-            dest=f"{prefix_with_dot.replace('-', '_')}STA_mode",
-            default=PipelineConfig.STA_mode.value,
-            choices=[mode.value for mode in STA_Mode],
-            help=
-            "STA mode: STA_inference, STA_searching, STA_tuning, STA_tuning_cfg, None",
-        )
-        parser.add_argument(
-            f"--{prefix_with_dot}skip-time-steps",
-            type=int,
-            dest=f"{prefix_with_dot.replace('-', '_')}skip_time_steps",
-            default=PipelineConfig.skip_time_steps,
-            help="Number of time steps to warmup (full attention) for STA",
-        )
-
        # Add VAE configuration arguments
        from fastvideo.configs.models.vaes.base import VAEConfig
        VAEConfig.add_cli_args(parser, prefix=f"{prefix_with_dot}vae-config")
@@ -266,7 +245,8 @@ class PipelineConfig:
        """
        use the pipeline class setting from model_path to match the pipeline config
        """
-        from fastvideo.registry import get_pipeline_config_cls_from_name
+        from fastvideo.configs.pipelines.registry import (
+            get_pipeline_config_cls_from_name)
        pipeline_config_cls = get_pipeline_config_cls_from_name(model_path)

        return cast(PipelineConfig, pipeline_config_cls(model_path=model_path))
@@ -280,7 +260,8 @@ class PipelineConfig:
        kwargs: dictionary of kwargs
        config_cli_prefix: prefix of CLI arguments for this PipelineConfig instance
        """
-        from fastvideo.registry import get_pipeline_config_cls_from_name
+        from fastvideo.configs.pipelines.registry import (
+            get_pipeline_config_cls_from_name)

        prefix_with_dot = f"{config_cli_prefix}." if (config_cli_prefix.strip()
                                                      != "") else ""
@@ -317,12 +298,6 @@ class PipelineConfig:
        # 4. Update PipelineConfig from CLI arguments if provided
        kwargs[prefix_with_dot + 'model_path'] = model_path
        pipeline_config.update_config_from_dict(kwargs, config_cli_prefix)
-
-        # Convert STA_mode string to enum if necessary
-        if isinstance(pipeline_config.STA_mode, str) and not isinstance(
-                pipeline_config.STA_mode, STA_Mode):
-            pipeline_config.STA_mode = STA_Mode(pipeline_config.STA_mode)
-
        return pipeline_config

    def check_pipeline_config(self) -> None:
@@ -11,8 +11,7 @@ from fastvideo.configs.models.dits import HunyuanVideo15Config
 from fastvideo.configs.models.encoders import (BaseEncoderOutput,
                                               Qwen2_5_VLConfig, T5Config)
 from fastvideo.configs.models.vaes import Hunyuan15VAEConfig
-from fastvideo.configs.models.upsamplers import SRTo720pUpsamplerConfig, SRTo1080pUpsamplerConfig
-from fastvideo.configs.pipelines.base import PipelineConfig, UpsamplerConfig
+from fastvideo.configs.pipelines.base import PipelineConfig

 PROMPT_TEMPLATE_TOKEN_LENGTH = 108

@@ -128,12 +127,12 @@ class Hunyuan15T2V480PConfig(PipelineConfig):
    vae_tiling: bool = True

    def __post_init__(self):
-        self.vae_config.load_encoder = False
+        self.vae_config.load_encoder = True
        self.vae_config.load_decoder = True


@dataclass
-class Hunyuan15I2V480PStepDistilledConfig(Hunyuan15T2V480PConfig):
+class Hunyuan15DistilledI2V480PConfig(Hunyuan15T2V480PConfig):
    flow_shift: int = 7


@@ -146,21 +145,11 @@ class Hunyuan15T2V720PConfig(Hunyuan15T2V480PConfig):


@dataclass
-class Hunyuan15I2V720PConfig(Hunyuan15T2V720PConfig):
-    """Base configuration for HunYuan pipeline architecture."""
-
-    # HunyuanConfig-specific parameters with defaults
-    flow_shift: int = 7
-
-
-@dataclass
-class Hunyuan15SR1080PConfig(Hunyuan15T2V720PConfig):
-    """Base configuration for HunYuan pipeline architecture."""
-
-    # HunyuanConfig-specific parameters with defaults
-    flow_shift: int = 7
-    flow_shift_sr: int = 2
-    upsampler_config: tuple[UpsamplerConfig, ...] = field(
-        default_factory=lambda:
-        (SRTo720pUpsamplerConfig(), SRTo1080pUpsamplerConfig()))
-    upsampler_precision: str = "fp32"
+class SelfForcingHunyuan15T2V480PConfig(Hunyuan15T2V480PConfig):
+    flow_shift: int = 5
+    is_causal: bool = True
+    # dmd_denoising_steps: list[int] | None = field(
+    #     default_factory=lambda: [1000, 750, 500, 250])
+    dmd_denoising_steps: list[int] | None = field(
+        default_factory=lambda: [1000, 875, 750, 625, 500, 375, 250, 125])
+    warp_denoising_step: bool = True
@@ -1,122 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-"""
-Pipeline configuration for HunyuanGameCraft.
-
-HunyuanGameCraft extends HunyuanVideo with:
-1. CameraNet for camera/action conditioning (Plücker coordinates)
-2. Mask-based conditioning for autoregressive generation
-3. 33 input channels (16 latent + 16 gt_latent + 1 mask)
-
-Text encoders are the same as HunyuanVideo:
- LLaVA-LLaMA-3-8B for primary text encoding (4096 dim)
- CLIP ViT-L/14 for secondary pooled embeddings (768 dim)
-"""
-from collections.abc import Callable
-from dataclasses import dataclass, field
-from typing import TypedDict
-
-import torch
-
-from fastvideo.configs.models import DiTConfig, EncoderConfig, VAEConfig
-from fastvideo.configs.models.dits import HunyuanGameCraftConfig
-from fastvideo.configs.models.encoders import (
-    BaseEncoderOutput,
-    CLIPTextConfig,
-    LlamaConfig,
-)
-from fastvideo.configs.models.vaes import GameCraftVAEConfig
-from fastvideo.configs.pipelines.base import PipelineConfig
-
-# GameCraft uses the same prompt template as HunyuanVideo
-PROMPT_TEMPLATE_ENCODE_VIDEO = (
-    "<|start_header_id|>system<|end_header_id|>\n\nDescribe the video by detailing the following aspects: "
-    "1. The main content and theme of the video."
-    "2. The color, shape, size, texture, quantity, text, and spatial relationships of the objects."
-    "3. Actions, events, behaviors temporal relationships, physical movement changes of the objects."
-    "4. background environment, light, style and atmosphere."
-    "5. camera angles, movements, and transitions used in the video:<|eot_id|>"
-    "<|start_header_id|>user<|end_header_id|>\n\n{}<|eot_id|>")
-
-
-class PromptTemplate(TypedDict):
-    template: str
-    crop_start: int
-
-
-prompt_template_video: PromptTemplate = {
-    "template": PROMPT_TEMPLATE_ENCODE_VIDEO,
-    "crop_start": 95,
-}
-
-
-def llama_preprocess_text(prompt: str) -> str:
-    """Apply prompt template for LLaMA encoder."""
-    return prompt_template_video["template"].format(prompt)
-
-
-def llama_postprocess_text(outputs: BaseEncoderOutput) -> torch.Tensor:
-    """Extract hidden states from LLaMA output, skipping instruction tokens."""
-    hidden_state_skip_layer = 2
-    assert outputs.hidden_states is not None
-    hidden_states: tuple[torch.Tensor, ...] = outputs.hidden_states
-    last_hidden_state: torch.Tensor = hidden_states[-(hidden_state_skip_layer +
-                                                      1)]
-    crop_start = prompt_template_video.get("crop_start", -1)
-    last_hidden_state = last_hidden_state[:, crop_start:]
-    return last_hidden_state
-
-
-def clip_preprocess_text(prompt: str) -> str:
-    """No preprocessing for CLIP encoder."""
-    return prompt
-
-
-def clip_postprocess_text(outputs: BaseEncoderOutput) -> torch.Tensor:
-    """Extract pooled output from CLIP encoder."""
-    pooler_output: torch.Tensor = outputs.pooler_output
-    return pooler_output
-
-
-@dataclass
-class HunyuanGameCraftPipelineConfig(PipelineConfig):
-    """Configuration for HunyuanGameCraft pipeline.
-    
-    Inherits text encoding from HunyuanVideo but uses:
-    - GameCraft DiT with CameraNet
-    - Same VAE (HunyuanVAE)
-    - Same text encoders (LLaMA + CLIP)
-    """
-
-    # DiT config - uses GameCraft config (33 input channels)
-    dit_config: DiTConfig = field(default_factory=HunyuanGameCraftConfig)
-
-    # VAE config - GameCraft VAE (mid_block_causal_attn=True, etc.)
-    vae_config: VAEConfig = field(default_factory=GameCraftVAEConfig)
-
-    # Denoising parameters
-    # Official GameCraft does NOT use embedded guidance (passes guidance=None)
-    # It uses standard CFG with guidance_scale=6.0 instead
-    embedded_cfg_scale = None
-    flow_shift: int = 5  # Official GameCraft uses flow_shift=5.0
-
-    # Text encoding stage - same as HunyuanVideo
-    # Uses LLaMA-3-8B (via LLaVA) + CLIP
-    text_encoder_configs: tuple[EncoderConfig, ...] = field(
-        default_factory=lambda: (LlamaConfig(), CLIPTextConfig()))
-    preprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
-        default_factory=lambda: (llama_preprocess_text, clip_preprocess_text))
-    postprocess_text_funcs: tuple[
-        Callable[[BaseEncoderOutput], torch.Tensor],
-        ...] = field(default_factory=lambda:
-                     (llama_postprocess_text, clip_postprocess_text))
-
-    # Precision for each component
-    dit_precision: str = "bf16"
-    vae_precision: str = "fp16"
-    text_encoder_precisions: tuple[str, ...] = field(
-        default_factory=lambda: ("fp16", "fp16"))
-
-    def __post_init__(self):
-        # VAE only needs decoder for inference
-        self.vae_config.load_encoder = False
-        self.vae_config.load_decoder = True
@@ -1,13 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-from dataclasses import dataclass, field
-
-from fastvideo.configs.models import DiTConfig
-from fastvideo.configs.pipelines.wan import Wan2_2_I2V_A14B_Config
-from fastvideo.configs.models.dits.lingbotworld import LingBotWorldVideoConfig
-
-
-@dataclass
-class LingBotWorldI2V480PConfig(Wan2_2_I2V_A14B_Config):
-    dit_config: DiTConfig = field(default_factory=LingBotWorldVideoConfig)
-    flow_shift: float | None = 10.0
-    boundary_ratio: float | None = 0.947
@@ -0,0 +1,218 @@
+# SPDX-License-Identifier: Apache-2.0
+"""Registry for pipeline weight-specific configurations."""
+
+import os
+from collections.abc import Callable
+
+from fastvideo.configs.pipelines.base import PipelineConfig
+from fastvideo.configs.pipelines.cosmos import CosmosConfig
+from fastvideo.configs.pipelines.cosmos2_5 import Cosmos25Config
+from fastvideo.configs.pipelines.hunyuan import FastHunyuanConfig, HunyuanConfig
+from fastvideo.configs.pipelines.hyworld import HYWorldConfig
+from fastvideo.configs.pipelines.ltx2 import LTX2T2VConfig
+from fastvideo.configs.pipelines.hunyuan15 import Hunyuan15T2V480PConfig, Hunyuan15T2V720PConfig, SelfForcingHunyuan15T2V480PConfig, Hunyuan15DistilledI2V480PConfig
+from fastvideo.configs.pipelines.stepvideo import StepVideoT2VConfig
+from fastvideo.configs.pipelines.longcat import LongCatT2V480PConfig
+from fastvideo.configs.pipelines.turbodiffusion import (
+    TurboDiffusionT2V_1_3B_Config, TurboDiffusionT2V_14B_Config,
+    TurboDiffusionI2V_A14B_Config)
+
+# isort: off
+from fastvideo.configs.pipelines.wan import (
+    FastWan2_1_T2V_480P_Config, FastWan2_2_TI2V_5B_Config,
+    Wan2_2_I2V_A14B_Config, Wan2_2_T2V_A14B_Config, Wan2_2_TI2V_5B_Config,
+    WanI2V480PConfig, WanI2V720PConfig, WanT2V480PConfig, WanT2V720PConfig,
+    SelfForcingWanT2V480PConfig, WANV2VConfig, SelfForcingWan2_2_T2V480PConfig,
+    MatrixGameI2V480PConfig)
+# isort: on
+from fastvideo.logger import init_logger
+from fastvideo.utils import (maybe_download_model_index,
+                             verify_model_config_and_directory)
+
+logger = init_logger(__name__)
+
+# Registry maps specific model weights to their config classes
+PIPE_NAME_TO_CONFIG: dict[str, type[PipelineConfig]] = {
+    "FastVideo/FastHunyuan-diffusers": FastHunyuanConfig,
+    "hunyuanvideo-community/HunyuanVideo": HunyuanConfig,
+    "weizhou03/SFHunyuanVideo-1.5-Diffusers-480p_t2v":
+    SelfForcingHunyuan15T2V480PConfig,
+    "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_i2v_step_distilled":
+    Hunyuan15DistilledI2V480PConfig,
+    "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v":
+    Hunyuan15T2V480PConfig,
+    "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v":
+    Hunyuan15T2V720PConfig,
+    "FastVideo/HY-WorldPlay-Bidirectional-Diffusers": HYWorldConfig,
+    "Wan-AI/Wan2.1-T2V-1.3B-Diffusers": WanT2V480PConfig,
+    "weizhou03/Wan2.1-Fun-1.3B-InP-Diffusers": WanI2V480PConfig,
+    "IRMChen/Wan2.1-Fun-1.3B-Control-Diffusers": WANV2VConfig,
+    "Wan-AI/Wan2.1-I2V-14B-480P-Diffusers": WanI2V480PConfig,
+    "Wan-AI/Wan2.1-I2V-14B-720P-Diffusers": WanI2V720PConfig,
+    "Wan-AI/Wan2.1-T2V-14B-Diffusers": WanT2V720PConfig,
+    "FastVideo/FastWan2.1-T2V-1.3B-Diffusers": FastWan2_1_T2V_480P_Config,
+    "FastVideo/FastWan2.1-T2V-14B-480P-Diffusers": FastWan2_1_T2V_480P_Config,
+    "FastVideo/FastWan2.2-TI2V-5B-Diffusers": FastWan2_2_TI2V_5B_Config,
+    "FastVideo/stepvideo-t2v-diffusers": StepVideoT2VConfig,
+    "FastVideo/Wan2.1-VSA-T2V-14B-720P-Diffusers": WanT2V720PConfig,
+    "wlsaidhi/SFWan2.1-T2V-1.3B-Diffusers": SelfForcingWanT2V480PConfig,
+    "rand0nmr/SFWan2.2-T2V-A14B-Diffusers": SelfForcingWan2_2_T2V480PConfig,
+    "FastVideo/SFWan2.2-I2V-A14B-Preview-Diffusers":
+    SelfForcingWan2_2_T2V480PConfig,
+    "Wan-AI/Wan2.2-TI2V-5B-Diffusers": Wan2_2_TI2V_5B_Config,
+    "Wan-AI/Wan2.2-T2V-A14B-Diffusers": Wan2_2_T2V_A14B_Config,
+    "Wan-AI/Wan2.2-I2V-A14B-Diffusers": Wan2_2_I2V_A14B_Config,
+    "nvidia/Cosmos-Predict2-2B-Video2World": CosmosConfig,
+    "KyleShao/Cosmos-Predict2.5-2B-Diffusers": Cosmos25Config,
+    "FastVideo/Matrix-Game-2.0-Base-Diffusers": MatrixGameI2V480PConfig,
+    "FastVideo/Matrix-Game-2.0-GTA-Diffusers": MatrixGameI2V480PConfig,
+    "FastVideo/Matrix-Game-2.0-TempleRun-Diffusers": MatrixGameI2V480PConfig,
+    # LongCat Video models
+    "FastVideo/LongCat-Video-T2V-Diffusers": LongCatT2V480PConfig,
+    "FastVideo/LongCat-Video-I2V-Diffusers": LongCatT2V480PConfig,
+    "FastVideo/LongCat-Video-VC-Diffusers": LongCatT2V480PConfig,
+    # LTX-2 models
+    "Lightricks/LTX-2": LTX2T2VConfig,
+    "converted/ltx2_diffusers": LTX2T2VConfig,
+    # TurboDiffusion models
+    "loayrashid/TurboWan2.1-T2V-1.3B-Diffusers": TurboDiffusionT2V_1_3B_Config,
+    "loayrashid/TurboWan2.1-T2V-14B-Diffusers": TurboDiffusionT2V_14B_Config,
+    "loayrashid/TurboWan2.2-I2V-A14B-Diffusers": TurboDiffusionI2V_A14B_Config,
+    # Add other specific weight variants
+}
+
+# For determining pipeline type from model ID
+PIPELINE_DETECTOR: dict[str, Callable[[str], bool]] = {
+    "longcatimagetovideo":
+    lambda id: "longcatimagetovideo" in id.lower(),
+    "longcatvideocontinuation":
+    lambda id: "longcatvideocontinuation" in id.lower(),
+    "longcat":
+    lambda id: "longcat" in id.lower(),
+    "hunyuan":
+    lambda id: "hunyuan" in id.lower(),
+    "hunyuan15":
+    lambda id: "hunyuan15" in id.lower(),
+    "hyworld":
+    lambda id: "hyworld" in id.lower(),
+    "matrixgame":
+    lambda id: "matrix-game" in id.lower() or "matrixgame" in id.lower(),
+    "wanpipeline":
+    lambda id: "wanpipeline" in id.lower(),
+    "wanimagetovideo":
+    lambda id: "wanimagetovideo" in id.lower(),
+    "wandmdpipeline":
+    lambda id: "wandmdpipeline" in id.lower(),
+    "wancausaldmdpipeline":
+    lambda id: "wancausaldmdpipeline" in id.lower(),
+    "stepvideo":
+    lambda id: "stepvideo" in id.lower(),
+    "cosmos":
+    lambda id: "cosmos" in id.lower() and ("2.5" not in id.lower(
+    ) and "2_5" not in id.lower() and "25" not in id.lower()),
+    "cosmos25":
+    lambda id: "cosmos25" in id.lower(),
+    "turbodiffusion":
+    lambda id: "turbodiffusion" in id.lower() or "turbowan" in id.lower(),
+    "ltx2":
+    lambda id: "ltx2" in id.lower() or "ltx-2" in id.lower(),
+    # Add other pipeline architecture detectors
+}
+
+# Fallback configs when exact match isn't found but architecture is detected
+PIPELINE_FALLBACK_CONFIG: dict[str, type[PipelineConfig]] = {
+    "longcatimagetovideo": LongCatT2V480PConfig,
+    "longcatvideocontinuation": LongCatT2V480PConfig,
+    "longcat": LongCatT2V480PConfig,
+    "cosmos25": Cosmos25Config,
+    "hunyuan":
+    HunyuanConfig,  # Base Hunyuan config as fallback for any Hunyuan variant
+    "matrixgame": MatrixGameI2V480PConfig,
+    "hunyuan15":
+    Hunyuan15T2V480PConfig,  # Base Hunyuan15 config as fallback for any Hunyuan15 variant
+    "hyworld":
+    HYWorldConfig,  # HYWorld-specific config as fallback for any HYWorld variant
+    "wanpipeline":
+    WanT2V480PConfig,  # Base Wan config as fallback for any Wan variant
+    "wanimagetovideo": WanI2V480PConfig,
+    "wandmdpipeline": FastWan2_1_T2V_480P_Config,
+    "wancausaldmdpipeline": SelfForcingWanT2V480PConfig,
+    "stepvideo": StepVideoT2VConfig,
+    "turbodiffusion": TurboDiffusionT2V_1_3B_Config,
+    "ltx2": LTX2T2VConfig,
+    # Other fallbacks by architecture
+}
+
+
+def get_pipeline_config_cls_from_name(
+        pipeline_name_or_path: str) -> type[PipelineConfig]:
+    """Get the appropriate configuration class for a given pipeline name or path.
+
+    This function implements a multi-step lookup process to find the most suitable
+    configuration class for a given pipeline. It follows this order:
+    1. Exact match in the PIPE_NAME_TO_CONFIG
+    2. Partial match in the PIPE_NAME_TO_CONFIG
+    3. Fallback to class name in the model_index.json
+    4. else raise an error
+
+    Args:
+        pipeline_name_or_path (str): The name or path of the pipeline. This can be:
+            - A registered model ID (e.g., "FastVideo/FastHunyuan-diffusers")
+            - A local path to a model directory
+            - A model ID that will be downloaded
+
+    Returns:
+        Type[PipelineConfig]: The configuration class that best matches the pipeline.
+            This will be one of:
+            - A specific weight configuration class if an exact match is found
+            - A fallback configuration class based on the pipeline architecture
+            - The base PipelineConfig class if no matches are found
+
+    Note:
+        - For local paths, the function will verify the model configuration
+        - For remote models, it will attempt to download the model index
+        - Warning messages are logged when falling back to less specific configurations
+    """
+
+    pipeline_config_cls: type[PipelineConfig] | None = None
+
+    # First try exact match for specific weights
+    if pipeline_name_or_path in PIPE_NAME_TO_CONFIG:
+        pipeline_config_cls = PIPE_NAME_TO_CONFIG[pipeline_name_or_path]
+        return pipeline_config_cls
+
+    # Try partial matches (for local paths that might include the weight ID)
+    for registered_id, config_class in PIPE_NAME_TO_CONFIG.items():
+        if registered_id in pipeline_name_or_path:
+            pipeline_config_cls = config_class
+            break
+
+    # If no match, try to use the fallback config
+    if pipeline_config_cls is None:
+        if os.path.exists(pipeline_name_or_path):
+            config = verify_model_config_and_directory(pipeline_name_or_path)
+        else:
+            config = maybe_download_model_index(pipeline_name_or_path)
+        logger.warning(
+            "Trying to use the config from the model_index.json. FastVideo may not correctly identify the optimal config for this model in this situation."
+        )
+
+        pipeline_name = config["_class_name"]
+        # Try to determine pipeline architecture for fallback
+        for pipeline_type, detector in PIPELINE_DETECTOR.items():
+            if detector(pipeline_name.lower()):
+                pipeline_config_cls = PIPELINE_FALLBACK_CONFIG.get(
+                    pipeline_type)
+                break
+
+        if pipeline_config_cls is not None:
+            logger.warning(
+                "No match found for pipeline %s, using fallback config %s.",
+                pipeline_name_or_path, pipeline_config_cls)
+
+    if pipeline_config_cls is None:
+        raise ValueError(
+            f"No match found for pipeline {pipeline_name_or_path}, please check the pipeline name or path."
+        )
+
+    return pipeline_config_cls
@@ -1,65 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-
-from __future__ import annotations
-
-from collections.abc import Callable
-from dataclasses import dataclass, field
-
-import torch
-
-from fastvideo.configs.models import EncoderConfig
-from fastvideo.configs.models.encoders import (
-    BaseEncoderOutput,
-    CLIPTextConfig,
-    T5Config,
-)
-from fastvideo.configs.models.dits.sd3 import SD3DiTConfig
-from fastvideo.configs.models.vaes.autoencoder_kl import AutoencoderKLVAEConfig
-from fastvideo.configs.pipelines.base import PipelineConfig, preprocess_text
-
-
-def _sd35_text_postprocess(outputs: BaseEncoderOutput) -> torch.Tensor:
-    assert outputs.last_hidden_state is not None
-    return outputs.last_hidden_state
-
-
-@dataclass
-class SD35Config(PipelineConfig):
-
-    scheduler_arch: str = "FlowMatchEulerDiscreteScheduler"
-    transformer_arch: str = "SD3Transformer2DModel"
-    vae_arch: str = "AutoencoderKL"
-    text_encoder_archs: tuple[str, ...] = (
-        "CLIPTextModelWithProjection",
-        "CLIPTextModelWithProjection",
-        "T5EncoderModel",
-    )
-    tokenizer_archs: tuple[str, ...] = (
-        "CLIPTokenizer",
-        "CLIPTokenizer",
-        "T5TokenizerFast",
-    )
-
-    dit_config: SD3DiTConfig = field(default_factory=SD3DiTConfig)
-    vae_config: AutoencoderKLVAEConfig = field(
-        default_factory=AutoencoderKLVAEConfig)
-
-    embedded_cfg_scale: float = 0.0
-    flow_shift: float | None = None
-
-    text_encoder_configs: tuple[EncoderConfig, ...] = field(
-        default_factory=lambda:
-        (CLIPTextConfig(), CLIPTextConfig(), T5Config()))
-    preprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
-        default_factory=lambda:
-        (preprocess_text, preprocess_text, preprocess_text))
-    postprocess_text_funcs: tuple[
-        Callable[[BaseEncoderOutput], torch.Tensor],
-        ...] = field(default_factory=lambda:
-                     (_sd35_text_postprocess, _sd35_text_postprocess,
-                      _sd35_text_postprocess))
-
-    dit_precision: str = "bf16"
-    vae_precision: str = "fp32"
-    text_encoder_precisions: tuple[str, ...] = field(
-        default_factory=lambda: ("fp32", "fp32", "bf16"))
@@ -41,7 +41,7 @@ class WanT2V480PConfig(PipelineConfig):
    vae_sp: bool = False

    # Denoising stage
-    flow_shift: float | None = 3.0
+    flow_shift: float | None = 5.0

    # Text encoding stage
    text_encoder_configs: tuple[EncoderConfig, ...] = field(
@@ -176,6 +176,8 @@ class SelfForcingWanT2V480PConfig(WanT2V480PConfig):
    flow_shift: float | None = 5.0
    dmd_denoising_steps: list[int] | None = field(
        default_factory=lambda: [1000, 750, 500, 250])
+    # dmd_denoising_steps: list[int] | None = field(
+    #     default_factory=lambda: [1000, 680, 360, 180])
    warp_denoising_step: bool = True


@@ -1,13 +1,3 @@
 from fastvideo.configs.sample.base import SamplingParam
-from fastvideo.configs.sample.hunyuangamecraft import (
-    HunyuanGameCraftSamplingParam,
-    HunyuanGameCraft65FrameSamplingParam,
-    HunyuanGameCraft129FrameSamplingParam,
-)

-__all__ = [
-    "SamplingParam",
-    "HunyuanGameCraftSamplingParam",
-    "HunyuanGameCraft65FrameSamplingParam",
-    "HunyuanGameCraft129FrameSamplingParam",
-]
+__all__ = ["SamplingParam"]
@@ -28,12 +28,6 @@ class SamplingParam:
    keyboard_cond: Any | None = None  # Shape: (B, T, K)
    grid_sizes: Any | None = None  # Shape: (3,) [F,H,W]

-    # Camera control inputs (HYWorld)
-    pose: str | None = None  # Camera trajectory: pose string (e.g., 'w-31') or JSON file path
-
-    # Camera control inputs (LingBotWorld)
-    c2ws_plucker_emb: Any | None = None  # Plucker embedding: [B, C, F_lat, H_lat, W_lat]
-
    # Refine inputs (LongCat 480p->720p upscaling)
    # Path-based refine (load stage1 video from disk, e.g. MP4)
    refine_from: str | None = None  # Path to stage1 video (480p output from distill)
@@ -61,13 +55,10 @@ class SamplingParam:
    num_frames_round_down: bool = False  # Whether to round down num_frames if it's not divisible by num_gpus
    height: int = 720
    width: int = 1280
-    height_sr: int = 1072
-    width_sr: int = 1920
    fps: int = 24

    # Denoising parameters
    num_inference_steps: int = 50
-    num_inference_steps_sr: int = 50
    guidance_scale: float = 1.0
    guidance_rescale: float = 0.0
    boundary_ratio: float | None = None
@@ -101,7 +92,8 @@ class SamplingParam:

    @classmethod
    def from_pretrained(cls, model_path: str) -> "SamplingParam":
-        from fastvideo.registry import get_sampling_param_cls_for_name
+        from fastvideo.configs.sample.registry import (
+            get_sampling_param_cls_for_name)
        sampling_cls = get_sampling_param_cls_for_name(model_path)
        if sampling_cls is not None:
            sampling_param: SamplingParam = sampling_cls()
@@ -5,19 +5,15 @@ from fastvideo.configs.sample.base import SamplingParam


@dataclass
-class Cosmos25SamplingParamBase(SamplingParam):
-    height: int = 704
-    width: int = 1280
-    num_frames: int = 77
+class Cosmos_Predict2_5_2B_Diffusers_SamplingParam(SamplingParam):
+    """Defaults for Cosmos 2.5 (Predict2.5) text-to-video diffusers-format model."""
+
+    height: int = 480
+    width: int = 832
+    num_frames: int = 121
    fps: int = 24
-    seed: int = 0

    guidance_scale: float = 7.0
-    negative_prompt: str = (
-        "The video captures a series of frames showing ugly scenes, static with no motion, motion blur, "
-        "over-saturation, shaky footage, low resolution, grainy texture, pixelated images, poorly lit areas, "
-        "underexposed and overexposed scenes, poor color balance, washed out colors, choppy sequences, jerky movements, "
-        "low frame rate, artifacting, color banding, unnatural transitions, outdated special effects, fake elements, "
-        "unconvincing visuals, poorly edited content, jump cuts, visual noise, and flickering. "
-        "Overall, the video is of poor quality.")
+    # Official Cosmos2.5 sampling uses empty string as unconditional.
+    negative_prompt: str = ""
    num_inference_steps: int = 35
@@ -17,8 +17,8 @@ class Hunyuan15_480P_SamplingParam(SamplingParam):
    guidance_scale: float = 6.0
    prompt_attention_mask: list = field(default_factory=list)
    negative_attention_mask: list = field(default_factory=list)
-    sigmas: list[float] | None = field(
-        default_factory=lambda: list(np.linspace(1.0, 0.0, 50 + 1)[:-1]))
+    # sigmas: list[float] | None = field(
+    #     default_factory=lambda: list(np.linspace(1.0, 0.0, 50 + 1)[:-1]))

    negative_prompt: str = ""

@@ -29,12 +29,10 @@ class Hunyuan15_480P_SamplingParam(SamplingParam):


@dataclass
-class Hunyuan15_480P_StepDistilled_I2V_SamplingParam(
-        Hunyuan15_480P_SamplingParam):
-    num_inference_steps: int = 12
+class Hunyuan15_480P_Distilled_SamplingParam(Hunyuan15_480P_SamplingParam):
+    num_inference_steps: int = 8
+    fps: int = 24

-    height: int = 720
-    width: int = 1280
    guidance_scale: float = 1.0


@@ -42,19 +40,3 @@ class Hunyuan15_480P_StepDistilled_I2V_SamplingParam(
 class Hunyuan15_720P_SamplingParam(Hunyuan15_480P_SamplingParam):
    height: int = 720
    width: int = 1280
-
-
-@dataclass
-class Hunyuan15_720P_Distilled_I2V_SamplingParam(Hunyuan15_720P_SamplingParam):
-    guidance_scale: float = 1.0
-
-
-@dataclass
-class Hunyuan15_SR_1080P_SamplingParam(Hunyuan15_480P_SamplingParam):
-    height_sr: int = 1072
-    width_sr: int = 1920
-
-    num_inference_steps: int = 12
-    num_inference_steps_sr: int = 8
-
-    guidance_scale: float = 1.0
@@ -1,104 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-"""
-Sampling parameters for HunyuanGameCraft video generation.
-
-GameCraft generates game-like videos with camera/action control.
-Default parameters are based on the official implementation.
-"""
-from dataclasses import dataclass, field
-from typing import Any
-
-from fastvideo.configs.sample.base import SamplingParam
-from fastvideo.configs.sample.teacache import TeaCacheParams
-
-
-@dataclass
-class HunyuanGameCraftSamplingParam(SamplingParam):
-    """Sampling parameters for HunyuanGameCraft video generation.
-    
-    Supports camera/action conditioning via:
-    - camera_trajectory: Plücker coordinates for camera motion
-    - action_list: List of actions (e.g., ["forward", "left", "right"])
-    - action_speed_list: Speed multipliers for each action
-    
-    Default resolution is 704x1280 (same as HunyuanVideo).
-    Default frame count is 33 video frames -> 9 latent frames.
-    """
-
-    # Number of denoising steps
-    num_inference_steps: int = 50
-
-    # Video dimensions
-    # 33 video frames -> 9 latent frames (4x temporal compression)
-    num_frames: int = 33
-    height: int = 704
-    width: int = 1280
-    fps: int = 24
-
-    # Guidance scale - official GameCraft uses CFG with guidance_scale=6.0
-    guidance_scale: float = 6.0
-
-    # Negative prompt for CFG (empty string = unconditional)
-    negative_prompt: str = ""
-
-    # Camera/Action conditioning
-    # Camera states as Plücker coordinates [B, T_video, 6, H, W]
-    camera_states: Any | None = None
-
-    # Camera trajectory file/identifier (alternative to camera_states)
-    camera_trajectory: str | None = None
-
-    # Action list for camera motion (e.g., ["forward", "left"])
-    action_list: list[str] | None = None
-
-    # Speed multipliers for each action
-    action_speed_list: list[float] | None = None
-
-    # History frame conditioning (for autoregressive generation)
-    # Ground truth latents for conditioning [B, 16, T, H, W]
-    gt_latents: Any | None = None
-
-    # Mask for conditioning (1=use gt, 0=generate) [B, 1, T, H, W]
-    conditioning_mask: Any | None = None
-
-    # Number of conditioning frames (for autoregressive) - maps to num_cond_frames
-    num_cond_frames: int = 0
-
-    # TeaCache parameters (if enabled)
-    teacache_params: TeaCacheParams = field(
-        default_factory=lambda: TeaCacheParams(
-            teacache_thresh=0.15,
-            coefficients=[
-                7.33226126e+02, -4.01131952e+02, 6.75869174e+01,
-                -3.14987800e+00, 9.61237896e-02
-            ],
-        ))
-
-    def __post_init__(self) -> None:
-        super().__post_init__()
-        # Validate action lists
-        if (self.action_list is not None and self.action_speed_list is not None
-                and len(self.action_list) != len(self.action_speed_list)):
-            raise ValueError(
-                f"action_list length ({len(self.action_list)}) must match "
-                f"action_speed_list length ({len(self.action_speed_list)})")
-
-
-@dataclass
-class HunyuanGameCraft65FrameSamplingParam(HunyuanGameCraftSamplingParam):
-    """Sampling parameters for 65-frame GameCraft generation.
-    
-    65 video frames -> 17 latent frames (with first frame as key frame).
-    This is useful for longer video generation.
-    """
-    num_frames: int = 65
-
-
-@dataclass
-class HunyuanGameCraft129FrameSamplingParam(HunyuanGameCraftSamplingParam):
-    """Sampling parameters for 129-frame GameCraft generation.
-    
-    129 video frames -> 33 latent frames.
-    This is the maximum supported by the official implementation.
-    """
-    num_frames: int = 129
@@ -1,21 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-from dataclasses import dataclass
-from fastvideo.configs.sample.wan import Wan2_2_I2V_A14B_SamplingParam
-
-
-@dataclass
-class LingBotWorld_SamplingParam(Wan2_2_I2V_A14B_SamplingParam):
-    guidance_scale: float = 5.0  # high_noise
-    guidance_scale_2: float = 5.0  # low_noise
-    num_inference_steps: int = 70
-    boundary_ratio: float | None = 0.947
-    negative_prompt: str | None = (
-        "画面突变，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，"
-        "最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，"
-        "畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走，"
-        "镜头晃动，画面闪烁，模糊，噪点，水印，签名，文字，变形，扭曲，液化，不合逻辑的结构，卡顿，"
-        "PPT幻灯片感，过暗，欠曝，低对比度，霓虹灯光感，过度锐化，3D渲染感，人物，行人，游客，身体，"
-        "皮肤，肢体，面部特征，汽车，电线")
-    fps: int = 16
-    # NOTE(will): default boundary timestep is tracked by PipelineConfig, but
-    # can be overridden during sampling
@@ -5,51 +5,10 @@ from fastvideo.configs.sample.base import SamplingParam


@dataclass
-class LTX2BaseSamplingParam(SamplingParam):
-    """Default sampling parameters for LTX-2 base one-stage T2V.
-
-    Values follow the official LTX-2 one-stage defaults.
-    Multi-modal CFG params are read by ``LTX2DenoisingStage``.
+class LTX2SamplingParam(SamplingParam):
+    """Default sampling parameters for LTX-2 distilled T2V.
    """

-    seed: int = 10
-    num_frames: int = 121
-    height: int = 512
-    width: int = 768
-    fps: int = 24
-    num_inference_steps: int = 40
-    guidance_scale: float = 3.0
-    # Copied/following official LTX-2 DEFAULT_NEGATIVE_PROMPT.
-    negative_prompt: str = (
-        "blurry, out of focus, overexposed, underexposed, low contrast, "
-        "washed out colors, excessive noise, grainy texture, poor lighting, "
-        "flickering, motion blur, distorted proportions, unnatural skin "
-        "tones, deformed facial features, asymmetrical face, missing facial "
-        "features, extra limbs, disfigured hands, wrong hand count, "
-        "artifacts around text, inconsistent perspective, camera shake, "
-        "incorrect depth of field, background too sharp, background clutter, "
-        "distracting reflections, harsh shadows, inconsistent lighting "
-        "direction, color banding, cartoonish rendering, 3D CGI look, "
-        "unrealistic materials, uncanny valley effect, incorrect ethnicity, "
-        "wrong gender, exaggerated expressions, wrong gaze direction, "
-        "mismatched lip sync, silent or muted audio, distorted voice, "
-        "robotic voice, echo, background noise, off-sync audio, incorrect "
-        "dialogue, added dialogue, repetitive speech, jittery movement, "
-        "awkward pauses, incorrect timing, unnatural transitions, "
-        "inconsistent framing, tilted camera, flat lighting, inconsistent "
-        "tone, cinematic oversaturation, stylized filters, or AI artifacts.")
-    # Official LTX-2 multi-modal CFG defaults.
-    ltx2_cfg_scale_video: float = 3.0
-    ltx2_cfg_scale_audio: float = 7.0
-    ltx2_modality_scale_video: float = 3.0
-    ltx2_modality_scale_audio: float = 3.0
-    ltx2_rescale_scale: float = 0.7
-
-
-@dataclass
-class LTX2DistilledSamplingParam(SamplingParam):
-    """Default sampling parameters for LTX-2 distilled one-stage T2V."""
-
    seed: int = 10
    num_frames: int = 121
    height: int = 1024
@@ -59,7 +18,3 @@ class LTX2DistilledSamplingParam(SamplingParam):
    guidance_scale: float = 1.0
    # No default negative_prompt for distilled models
    negative_prompt: str = ""
-
-
-# Backward compatibility alias.
-LTX2SamplingParam = LTX2DistilledSamplingParam
@@ -0,0 +1,208 @@
+# SPDX-License-Identifier: Apache-2.0
+import os
+from collections.abc import Callable
+from typing import Any
+
+from fastvideo.configs.sample.hunyuan import (FastHunyuanSamplingParam,
+                                              HunyuanSamplingParam)
+from fastvideo.configs.sample.hyworld import HYWorld_SamplingParam
+from fastvideo.configs.sample.hunyuan15 import Hunyuan15_480P_SamplingParam, Hunyuan15_720P_SamplingParam, Hunyuan15_480P_Distilled_SamplingParam
+from fastvideo.configs.sample.stepvideo import StepVideoT2VSamplingParam
+
+from fastvideo.configs.sample.cosmos import Cosmos_Predict2_2B_Video2World_SamplingParam
+from fastvideo.configs.sample.cosmos2_5 import Cosmos_Predict2_5_2B_Diffusers_SamplingParam
+from fastvideo.configs.sample.ltx2 import LTX2SamplingParam
+
+# isort: off
+from fastvideo.configs.sample.wan import (
+    FastWanT2V480P_SamplingParam,
+    Wan2_1_Fun_1_3B_InP_SamplingParam,
+    Wan2_2_I2V_A14B_SamplingParam,
+    Wan2_2_T2V_A14B_SamplingParam,
+    Wan2_2_TI2V_5B_SamplingParam,
+    WanI2V_14B_480P_SamplingParam,
+    WanI2V_14B_720P_SamplingParam,
+    WanT2V_1_3B_SamplingParam,
+    WanT2V_14B_SamplingParam,
+    Wan2_1_Fun_1_3B_Control_SamplingParam,
+    SelfForcingWan2_1_T2V_1_3B_480P_SamplingParam,
+    SelfForcingWan2_2_T2V_A14B_480P_SamplingParam,
+    MatrixGame2_SamplingParam,
+)
+from fastvideo.configs.sample.turbodiffusion import (
+    TurboDiffusionT2V_1_3B_SamplingParam,
+    TurboDiffusionT2V_14B_SamplingParam,
+    TurboDiffusionI2V_A14B_SamplingParam,
+)
+# isort: on
+from fastvideo.logger import init_logger
+from fastvideo.utils import (maybe_download_model_index,
+                             verify_model_config_and_directory)
+
+logger = init_logger(__name__)
+# Registry maps specific model weights to their config classes
+SAMPLING_PARAM_REGISTRY: dict[str, Any] = {
+    "FastVideo/FastHunyuan-diffusers": FastHunyuanSamplingParam,
+    "hunyuanvideo-community/HunyuanVideo": HunyuanSamplingParam,
+    "weizhou03/SFHunyuanVideo-1.5-Diffusers-480p_t2v":
+    Hunyuan15_480P_SamplingParam,
+    "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_i2v_step_distilled":
+    Hunyuan15_480P_Distilled_SamplingParam,
+    "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v":
+    Hunyuan15_480P_SamplingParam,
+    "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v":
+    Hunyuan15_720P_SamplingParam,
+    "FastVideo/HY-WorldPlay-Bidirectional-Diffusers": HYWorld_SamplingParam,
+    "FastVideo/stepvideo-t2v-diffusers": StepVideoT2VSamplingParam,
+
+    # Wan2.1
+    "Wan-AI/Wan2.1-T2V-1.3B-Diffusers": WanT2V_1_3B_SamplingParam,
+    "Wan-AI/Wan2.1-T2V-14B-Diffusers": WanT2V_14B_SamplingParam,
+    "Wan-AI/Wan2.1-I2V-14B-480P-Diffusers": WanI2V_14B_480P_SamplingParam,
+    "Wan-AI/Wan2.1-I2V-14B-720P-Diffusers": WanI2V_14B_720P_SamplingParam,
+    "weizhou03/Wan2.1-Fun-1.3B-InP-Diffusers":
+    Wan2_1_Fun_1_3B_InP_SamplingParam,
+    "IRMChen/Wan2.1-Fun-1.3B-Control-Diffusers":
+    Wan2_1_Fun_1_3B_Control_SamplingParam,
+
+    # Wan2.2
+    "Wan-AI/Wan2.2-TI2V-5B-Diffusers": Wan2_2_TI2V_5B_SamplingParam,
+    "FastVideo/FastWan2.2-TI2V-5B-FullAttn-Diffusers":
+    Wan2_2_TI2V_5B_SamplingParam,
+    "Wan-AI/Wan2.2-T2V-A14B-Diffusers": Wan2_2_T2V_A14B_SamplingParam,
+    "Wan-AI/Wan2.2-I2V-A14B-Diffusers": Wan2_2_I2V_A14B_SamplingParam,
+
+    # FastWan2.1
+    "FastVideo/FastWan2.1-T2V-1.3B-Diffusers": FastWanT2V480P_SamplingParam,
+
+    # FastWan2.2
+    "FastVideo/FastWan2.2-TI2V-5B-Diffusers": Wan2_2_TI2V_5B_SamplingParam,
+
+    # Causal Self-Forcing Wan2.1
+    "wlsaidhi/SFWan2.1-T2V-1.3B-Diffusers":
+    SelfForcingWan2_1_T2V_1_3B_480P_SamplingParam,
+
+    # Causal Self-Forcing Wan2.2
+    "rand0nmr/SFWan2.2-T2V-A14B-Diffusers":
+    SelfForcingWan2_2_T2V_A14B_480P_SamplingParam,
+    "FastVideo/SFWan2.2-I2V-A14B-Preview-Diffusers":
+    SelfForcingWan2_2_T2V_A14B_480P_SamplingParam,
+
+    # Cosmos2
+    "nvidia/Cosmos-Predict2-2B-Video2World":
+    Cosmos_Predict2_2B_Video2World_SamplingParam,
+
+    # Cosmos2.5
+    "KyleShao/Cosmos-Predict2.5-2B-Diffusers":
+    Cosmos_Predict2_5_2B_Diffusers_SamplingParam,
+
+    # MatrixGame2.0 models
+    "FastVideo/Matrix-Game-2.0-Base-Diffusers": MatrixGame2_SamplingParam,
+    "FastVideo/Matrix-Game-2.0-GTA-Diffusers": MatrixGame2_SamplingParam,
+    "FastVideo/Matrix-Game-2.0-TempleRun-Diffusers": MatrixGame2_SamplingParam,
+
+    # TurboDiffusion models
+    "loayrashid/TurboWan2.1-T2V-1.3B-Diffusers":
+    TurboDiffusionT2V_1_3B_SamplingParam,
+    "loayrashid/TurboWan2.1-T2V-14B-Diffusers":
+    TurboDiffusionT2V_14B_SamplingParam,
+    "loayrashid/TurboWan2.2-I2V-A14B-Diffusers":
+    TurboDiffusionI2V_A14B_SamplingParam,
+
+    # LTX-2 models
+    "Lightricks/LTX-2": LTX2SamplingParam,
+    "FastVideo/LTX2-Distilled-Diffusers": LTX2SamplingParam,
+
+    # Add other specific weight variants
+}
+
+# For determining pipeline type from model ID
+SAMPLING_PARAM_DETECTOR: dict[str, Callable[[str], bool]] = {
+    "hunyuan":
+    lambda id: "hunyuan" in id.lower(),
+    "hunyuan15":
+    lambda id: "hunyuan15" in id.lower(),
+    "hyworld":
+    lambda id: "hyworld" in id.lower(),
+    "wanpipeline":
+    lambda id: "wanpipeline" in id.lower(),
+    "wanimagetovideo":
+    lambda id: "wanimagetovideo" in id.lower(),
+    "stepvideo":
+    lambda id: "stepvideo" in id.lower(),
+    "wandmdpipeline":
+    lambda id: "wandmdpipeline" in id.lower(),
+    "wancausaldmdpipeline":
+    lambda id: "wancausaldmdpipeline" in id.lower(),
+    "matrixgame":
+    lambda id: "matrixgame" in id.lower() or "matrix-game" in id.lower(),
+    "turbodiffusion":
+    lambda id: "turbodiffusion" in id.lower() or "turbowan" in id.lower(),
+    "cosmos25":
+    lambda id: "cosmos2_5" in id.lower(),
+    "cosmos":
+    lambda id: "cosmos" in id.lower() and "2_5" not in id.lower(),
+    "ltx2":
+    lambda id: "ltx2" in id.lower() or "ltx-2" in id.lower(),
+    # Add other pipeline architecture detectors
+}
+
+# Fallback configs when exact match isn't found but architecture is detected
+SAMPLING_FALLBACK_PARAM: dict[str, Any] = {
+    "hunyuan":
+    HunyuanSamplingParam,  # Base Hunyuan config as fallback for any Hunyuan variant
+    "hunyuan15":
+    Hunyuan15_480P_SamplingParam,  # Base Hunyuan15 config as fallback for any Hunyuan15 variant
+    "hyworld":
+    HYWorld_SamplingParam,  # HYWorld-specific config as fallback for any HYWorld variant
+    "wanpipeline":
+    WanT2V_1_3B_SamplingParam,  # Base Wan config as fallback for any Wan variant
+    "wanimagetovideo": WanI2V_14B_480P_SamplingParam,
+    "wandmdpipeline": FastWanT2V480P_SamplingParam,
+    "wancausaldmdpipeline": SelfForcingWan2_1_T2V_1_3B_480P_SamplingParam,
+    "stepvideo": StepVideoT2VSamplingParam,
+    "matrixgame": MatrixGame2_SamplingParam,
+    "turbodiffusion":
+    TurboDiffusionT2V_1_3B_SamplingParam,  # Default to T2V for fallback
+    "cosmos25": Cosmos_Predict2_5_2B_Diffusers_SamplingParam,
+    "cosmos": Cosmos_Predict2_2B_Video2World_SamplingParam,
+    "ltx2": LTX2SamplingParam,
+    # Other fallbacks by architecture
+}
+
+
+def get_sampling_param_cls_for_name(pipeline_name_or_path: str) -> Any | None:
+    """Get the appropriate sampling param for specific pretrained weights."""
+    # First try exact match for specific weights
+    if pipeline_name_or_path in SAMPLING_PARAM_REGISTRY:
+        return SAMPLING_PARAM_REGISTRY[pipeline_name_or_path]
+
+    # Try partial matches (for local paths that might include the weight ID)
+    for registered_id, config_class in SAMPLING_PARAM_REGISTRY.items():
+        if registered_id in pipeline_name_or_path:
+            return config_class
+
+    matrixgame_patterns = ["Matrix-Game", "Skywork--Matrix-Game", "matrixgame"]
+    for pattern in matrixgame_patterns:
+        if pattern.lower() in pipeline_name_or_path.lower():
+            return MatrixGame2_SamplingParam
+
+    if os.path.exists(pipeline_name_or_path):
+        config = verify_model_config_and_directory(pipeline_name_or_path)
+    else:
+        config = maybe_download_model_index(pipeline_name_or_path)
+
+    pipeline_name = config["_class_name"]
+
+    # If no match, try to use the fallback config
+    fallback_config = None
+    # Try to determine pipeline architecture for fallback
+    for pipeline_type, detector in SAMPLING_PARAM_DETECTOR.items():
+        if detector(pipeline_name.lower()):
+            fallback_config = SAMPLING_FALLBACK_PARAM.get(pipeline_type)
+            break
+
+    logger.warning(
+        "No match found for pipeline %s, using fallback sampling param %s.",
+        pipeline_name_or_path, fallback_config)
+    return fallback_config
@@ -1,25 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-
-from __future__ import annotations
-
-from dataclasses import dataclass
-
-from fastvideo.configs.sample.base import SamplingParam
-
-
-@dataclass
-class SD35SamplingParam(SamplingParam):
-
-    prompt: str | None = "a photo of a cat"
-    negative_prompt: str = ""
-
-    num_videos_per_prompt: int = 1
-    seed: int = 0
-
-    num_frames: int = 1
-    height: int = 512
-    width: int = 512
-    fps: int = 1
-
-    num_inference_steps: int = 28
-    guidance_scale: float = 6.0
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Wei Zhou	3c71218987	ckpt	2026-04-01 02:22:08 +00:00
JerryZhou54	6b4cc8772d	launched synthetic video generation for hy1.5	2026-02-08 23:46:16 +00:00
JerryZhou54	0df9934f4d	ckpt	2026-02-07 21:43:09 +00:00
JerryZhou54	8140cb4460	small fix	2026-01-28 19:10:28 +00:00
JerryZhou54	e42f688466	fix lint	2026-01-28 15:39:18 +00:00
JerryZhou54	4062953769	Modify distill scripts	2026-01-28 15:32:15 +00:00
JerryZhou54	8cebfdad67	Modify fastvideo/training	2026-01-28 15:32:15 +00:00
JerryZhou54	d5d061a1f7	revert hy15 preprocess related changes	2026-01-28 15:32:13 +00:00
JerryZhou54	07db3796d2	refactor hy15 sf distill	2026-01-28 15:30:21 +00:00
JerryZhou54	363d81f3bb	refactor hy15 causal denoising	2026-01-28 15:29:51 +00:00
JerryZhou54	2e0c5fa979	fix lint	2026-01-28 15:29:48 +00:00
JerryZhou54	448490b838	compatible with wan sf	2026-01-28 15:29:21 +00:00
JerryZhou54	5546fa9aad	Add context forcing, ode_init to sf training	2026-01-28 15:29:21 +00:00
JerryZhou54	9718cd974e	small change	2026-01-28 15:29:21 +00:00
JerryZhou54	8a623d5aa7	ckpt	2026-01-28 15:29:18 +00:00
JerryZhou54	d481af1d36	Ode init running for hy15	2026-01-28 15:28:26 +00:00
JerryZhou54	eda4148681	Ode runnable for hy15	2026-01-28 15:27:23 +00:00
JerryZhou54	9cf2a1561a	Add support for ode_init inference for hy15 & support multiple timesteps for hy15	2026-01-28 15:27:22 +00:00
				`@@ -0,0 +1 @@`
				`The raw mixkit dataset is downloaded and preprocessed following instructions from https://github.com/tianweiy/CausVid/tree/master`