Compare commits
| Author | SHA1 | Date | |
|---|---|---|---|
 William Lin
|
0ceff110da | ||
|
Yongqi Chen
|
1d018acb3e | ||
|
Yongqi Chen
|
7d8cf38dbe | ||
|
Yongqi Chen
|
8d483fe4aa | ||
|
Zhang Peiyuan
|
c1191250bf | ||
|
Wenxuan Tan
|
4b7266349a | ||
|
Yongqi Chen
|
22f9b7681f | ||
|
Yongqi Chen
|
589d32cc39 | ||
|
Hao Zhang
|
89199837db |
@@ -2,12 +2,12 @@
|
||||
<img src=assets/logo.png width="30%"/>
|
||||
</div>
|
||||
|
||||
**FastVideo is a unified framework for accelerated video generation.**
|
||||
**FastVideo is a unified post-training and inference framework for accelerated video generation.**
|
||||
|
||||
FastVideo is an inference and post-training framework for diffusion models. It features an end-to-end unified pipeline for accelerating diffusion models, starting from data preprocessing to model training, finetuning, distillation, and inference. FastVideo is designed to be modular and extensible, allowing users to easily add new optimizations and techniques. Whether it is training-free optimizations or post-training optimizations, FastVideo has you covered.
|
||||
FastVideo features an end-to-end unified pipeline for accelerating diffusion models, starting from data preprocessing to model training, finetuning, distillation, and inference. FastVideo is designed to be modular and extensible, allowing users to easily add new optimizations and techniques. Whether it is training-free optimizations or post-training optimizations, FastVideo has you covered.
|
||||
|
||||
<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.html"><b> Quick Start</b></a> | 🤗 <a href="https://huggingface.co/FastVideo/FastWan2.1-T2V-1.3B-Diffusers" target="_blank"><b>FastWan2.1</b></a> | 🤗 <a href="https://huggingface.co/FastVideo/FastWan2.2-TI2V-5B-Diffusers" target="_blank"><b>FastWan2.2</b></a> | 🟣💬 <a href="https://join.slack.com/t/fastvideo/shared_invite/zt-38u6p1jqe-yDI1QJOCEnbtkLoaI5bjZQ" target="_blank"> <b>Slack</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.html"><b> Quick Start</b></a> | 🤗 <a href="https://huggingface.co/FastVideo/FastWan2.1-T2V-1.3B-Diffusers" target="_blank"><b>FastWan2.1</b></a> | 🤗 <a href="https://huggingface.co/FastVideo/FastWan2.2-TI2V-5B-Diffusers" target="_blank"><b>FastWan2.2</b></a> | 🟣💬 <a href="https://join.slack.com/t/fastvideo/shared_invite/zt-38u6p1jqe-yDI1QJOCEnbtkLoaI5bjZQ" target="_blank"> <b>Slack</b> </a> | 🟣💬 <a href="https://ibb.co/qqPzbrw" target="_blank"> <b> WeChat </b> </a> |
|
||||
</p>
|
||||
|
||||
<div align="center">
|
||||
@@ -23,15 +23,19 @@ FastVideo is an inference and post-training framework for diffusion models. It f
|
||||
## Key Features
|
||||
|
||||
FastVideo has the following features:
|
||||
- End-to-end post-training support:
|
||||
- [Sparse distillation](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/) for Wan2.1 and Wan2.2 to achineve >50x denoising speedup
|
||||
- Data preprocessing pipeline for video data
|
||||
- Support full finetuning and LoRA finetuning for state-of-the-art open video DiTs
|
||||
- Scalable training with FSDP2, sequence parallelism, and selective activation checkpointing, with near linear scaling to 64 GPUs
|
||||
- State-of-the-art performance optimizations for inference
|
||||
- [Video Sparse Attention](https://arxiv.org/pdf/2505.13389)
|
||||
- [Sliding Tile Attention](https://arxiv.org/pdf/2502.04507)
|
||||
- [TeaCache](https://arxiv.org/pdf/2411.19108)
|
||||
- [Sage Attention](https://arxiv.org/abs/2410.02367)
|
||||
- E2E post-training support
|
||||
- Data preprocessing pipeline for video data.
|
||||
- [Sparse distillation](https://hao-ai-lab.github.io/blogs/fastvideo_post_training/) for Wan2.1 and Wan2.2 using [Video Sparse Attention](https://arxiv.org/pdf/2505.13389) and [Distribution Matching Distillation](https://tianweiy.github.io/dmd2/)
|
||||
- Support full finetuning and LoRA finetuning for state-of-the-art open video DiTs.
|
||||
- Scalable training with FSDP2, sequence parallelism, and selective activation checkpointing, with near linear scaling to 64 GPUs.
|
||||
- Diverse hardware and OS support
|
||||
- Support H100, A100, 4090
|
||||
- Support Linux, Windows, MacOS
|
||||
|
||||
## Getting Started
|
||||
We recommend using an environment manager such as `Conda` to create a clean environment:
|
||||
@@ -68,7 +72,7 @@ from fastvideo import VideoGenerator
|
||||
def main():
|
||||
# Create a video generator with a pre-trained model
|
||||
generator = VideoGenerator.from_pretrained(
|
||||
"Wan-AI/Wan2.1-T2V-1.3B-Diffusers",
|
||||
"FastVideo/FastWan2.1-T2V-1.3B-Diffusers",
|
||||
num_gpus=1, # Adjust based on your hardware
|
||||
)
|
||||
|
||||
@@ -105,18 +109,19 @@ For a more detailed guide, please see our [inference quick start](https://hao-ai
|
||||
<!-- - [Finetuning Guide](https://hao-ai-lab.github.io/FastVideo/training/finetune.html) -->
|
||||
|
||||
## 📑 Development Plan
|
||||
|
||||
<!-- - More distillation methods -->
|
||||
<!-- - [ ] Add Distribution Matching Distillation -->
|
||||
- More FastWan Models Coming Soon!
|
||||
- [ ] Add FastWan2.1-T2V-14B
|
||||
- [ ] Add FastWan2.2-T2V-14B
|
||||
- [ ] Add FastWan2.2-I2V-14B
|
||||
More FastWan Models Coming Soon!
|
||||
- [ ] Add FastWan2.1-T2V-14B
|
||||
- [ ] Add FastWan2.2-T2V-14B
|
||||
- [ ] Add FastWan2.2-I2V-14B
|
||||
<!-- - Optimization features
|
||||
- Code updates -->
|
||||
<!-- - [ ] fp8 support -->
|
||||
<!-- - [ ] faster load model and save model support -->
|
||||
|
||||
See details in [development roadmap](https://github.com/hao-ai-lab/FastVideo/issues/468).
|
||||
|
||||
## 🤝 Contributing
|
||||
|
||||
We welcome all contributions. Please check out our guide [here](https://hao-ai-lab.github.io/FastVideo/contributing/overview.html)
|
||||
@@ -132,10 +137,10 @@ We learned and reused code from the following projects:
|
||||
- [vLLM](https://github.com/vllm-project/vllm)
|
||||
- [SGLang](https://github.com/sgl-project/sglang)
|
||||
|
||||
We thank MBZUAI and [Anyscale](https://www.anyscale.com/) 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 use FastVideo for your research, please cite our work:
|
||||
If you find FastVideo useful, please considering citing our work:
|
||||
|
||||
```bibtex
|
||||
@software{fastvideo2024,
|
||||
@@ -146,22 +151,17 @@ If you use FastVideo for your research, please cite our work:
|
||||
year = {2024},
|
||||
}
|
||||
|
||||
@misc{zhang2025vsafastervideodiffusion,
|
||||
title={VSA: Faster Video Diffusion with Trainable Sparse Attention},
|
||||
author={Peiyuan Zhang and Haofeng Huang and Yongqi Chen and Will Lin and Zhengzhong Liu and Ion Stoica and Eric Xing and Hao Zhang},
|
||||
year={2025},
|
||||
eprint={2505.13389},
|
||||
archivePrefix={arXiv},
|
||||
primaryClass={cs.CV},
|
||||
url={https://arxiv.org/abs/2505.13389},
|
||||
@article{zhang2025vsa,
|
||||
title={VSA: Faster Video Diffusion with Trainable Sparse Attention},
|
||||
author={Zhang, Peiyuan and Huang, Haofeng and Chen, Yongqi and Lin, Will and Liu, Zhengzhong and Stoica, Ion and Xing, Eric and Zhang, Hao},
|
||||
journal={arXiv preprint arXiv:2505.13389},
|
||||
year={2025}
|
||||
}
|
||||
@misc{zhang2025fastvideogenerationsliding,
|
||||
title={Fast Video Generation with Sliding Tile Attention},
|
||||
author={Peiyuan Zhang and Yongqi Chen and Runlong Su and Hangliang Ding and Ion Stoica and Zhenghong Liu and Hao Zhang},
|
||||
year={2025},
|
||||
eprint={2502.04507},
|
||||
archivePrefix={arXiv},
|
||||
primaryClass={cs.CV},
|
||||
url={https://arxiv.org/abs/2502.04507},
|
||||
|
||||
@article{zhang2025fast,
|
||||
title={Fast video generation with sliding tile attention},
|
||||
author={Zhang, Peiyuan and Chen, Yongqi and Su, Runlong and Ding, Hangliang and Stoica, Ion and Liu, Zhengzhong and Zhang, Hao},
|
||||
journal={arXiv preprint arXiv:2502.04507},
|
||||
year={2025}
|
||||
}
|
||||
```
|
||||
|
||||
@@ -6,8 +6,9 @@ We introduce a new finetuning strategy - **Sparse-distill**, which jointly integ
|
||||
|
||||
We provide two distilled models:
|
||||
|
||||
- **[FastWan2.1-T2V-1.3B-Diffusers](https://huggingface.co/FastVideo/FastWan2.1-T2V-1.3B-Diffusers)**: 3-step inference, up to **20 FPS** on H100 GPU
|
||||
- **[FastWan2.1-T2V-14B-480P-Diffusers](https://huggingface.co/FastVideo/FastWan2.1-T2V-14B-480P-Diffusers)**: 3-step inference, up to **50x speed up** at 480P, **70x speed up** at 720P for denoising loop
|
||||
- **[FastWan2.1-T2V-1.3B-Diffusers](https://huggingface.co/FastVideo/FastWan2.1-T2V-1.3B-Diffusers)**: 3-step inference, up to **16 FPS** on H100 GPU
|
||||
- **[FastWan2.1-T2V-14B-480P-Diffusers](https://huggingface.co/FastVideo/FastWan2.1-T2V-14B-480P-Diffusers)**: 3-step inference, up to **60x speed up** at 480P, **90x speed up** at 720P for denoising loop
|
||||
- **[FastWan2.2-TI2V-5B-FullAttn-Diffusers](https://huggingface.co/FastVideo/FastWan2.2-TI2V-5B-FullAttn-Diffusers)**: 3-step inference, up to **50x speed up** at 720P for denoising loop
|
||||
|
||||
Both models are trained on **61×448×832** resolution but support generating videos with **any resolution** (1.3B model mainly support 480P, 14B model support 480P and 720P, quality may degrade for different resolutions).
|
||||
|
||||
@@ -34,7 +35,7 @@ python scripts/huggingface/download_hf.py \
|
||||
|
||||
## 🚀 Training Scripts
|
||||
|
||||
### 1.3B Model Sparse-Distill
|
||||
### Wan2.1 1.3B Model Sparse-Distill
|
||||
|
||||
For the 1.3B model, we use **4 nodes with 32 H200 GPUs** (8 GPUs per node):
|
||||
|
||||
@@ -50,7 +51,7 @@ sbatch examples/distill/Wan2.1-T2V/Wan-Syn-Data-480P/distill_dmd_VSA_t2v_1.3B.sl
|
||||
- VSA attention sparsity: 0.8
|
||||
- Training steps: 4000 (~12 hours)
|
||||
|
||||
### 14B Model Sparse-Distill
|
||||
### Wan2.1 14B Model Sparse-Distill
|
||||
|
||||
For the 14B model, we use **8 nodes with 64 H200 GPUs** (8 GPUs per node):
|
||||
|
||||
@@ -67,3 +68,20 @@ sbatch examples/distill/Wan2.1-T2V/Wan-Syn-Data-480P/distill_dmd_VSA_t2v_14B.slu
|
||||
- VSA attention sparsity: 0.9
|
||||
- Training steps: 3000 (~52 hours)
|
||||
- HSDP shard dim: 8
|
||||
|
||||
### Wan2.2 5B Model Sparse-Distill
|
||||
|
||||
For the 5B model, we use **8 nodes with 64 H200 GPUs** (8 GPUs per node):
|
||||
|
||||
```bash
|
||||
# Multi-node training (8 nodes, 64 GPUs total)
|
||||
sbatch examples/distill/Wan2.2-TI2V-5B-Diffusers/Data-free/distill_dmd_t2v_5B.sh
|
||||
```
|
||||
|
||||
**Key Configuration:**
|
||||
- Global batch size: 64
|
||||
- Sequence parallel size: 1
|
||||
- Gradient accumulation steps: 1
|
||||
- Learning rate: 2e-5
|
||||
- Training steps: 3000 (~12 hours)
|
||||
- HSDP shard dim: 1
|
||||
|
||||
@@ -9,7 +9,7 @@
|
||||
|
||||
:::{raw} html
|
||||
<p style="text-align:center">
|
||||
<strong>FastVideo is An unified inference and post-training framework for accelerated video generation.
|
||||
<strong>FastVideo is a unified inference and post-training framework for accelerated video generation.
|
||||
</strong>
|
||||
</p>
|
||||
|
||||
|
||||
@@ -53,3 +53,9 @@ out = sliding_tile_attention(q, k, v, window_size, text_length)
|
||||
out = sliding_tile_attention(q, k, v, window_size, 0, False)
|
||||
|
||||
```
|
||||
|
||||
# 🚀Inference
|
||||
|
||||
```bash
|
||||
bash scripts/inference/v1_inference_wan_STA.sh
|
||||
```
|
||||
|
||||
@@ -59,3 +59,9 @@ from vsa import video_sparse_attn
|
||||
output = video_sparse_attn(q, k, v, variable_block_sizes, topk, block_size, compress_attn_weight)
|
||||
|
||||
```
|
||||
|
||||
# 🚀Inference
|
||||
|
||||
```bash
|
||||
bash scripts/inference/v1_inference_wan_VSA.sh
|
||||
```
|
||||
|
||||
@@ -102,7 +102,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.0
|
||||
--dit_precision "fp32"
|
||||
|
||||
@@ -102,7 +102,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.0
|
||||
--dit_precision "fp32"
|
||||
|
||||
@@ -102,7 +102,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.0
|
||||
--dit_precision "fp32"
|
||||
|
||||
@@ -0,0 +1,13 @@
|
||||
# Wan2.2-5B Distill Example
|
||||
These are end-to-end example scripts for distilling Wan2.2 TI2V 5B model DMD+VSA methods.
|
||||
|
||||
### 0. Make sure you have installed VSA
|
||||
|
||||
```bash
|
||||
cd csrc/attn
|
||||
git submodule update --init --recursive
|
||||
python setup_vsa.py install
|
||||
```
|
||||
|
||||
### Data-free Distillation
|
||||
When `--simulate_generator_forward` is enabled, distillation becomes data-free by simulating intermediate steps through forward inference of the generator. This helps avoid training–inference mismatch. See Section 4.5 of [DMD2](https://arxiv.org/pdf/2405.14867) for details.
|
||||
@@ -108,7 +108,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.0
|
||||
--dit_precision "fp32"
|
||||
|
||||
@@ -108,7 +108,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.0
|
||||
--dit_precision "fp32"
|
||||
|
||||
@@ -9,4 +9,14 @@ git submodule update --init --recursive
|
||||
python setup_vsa.py install
|
||||
```
|
||||
|
||||
### TODO
|
||||
### 1. Download dataset:
|
||||
```bash
|
||||
bash examples/distill/Wan2.2-TI2V-5B-Diffusers/crush_smol/download_dataset.sh
|
||||
```
|
||||
|
||||
### 2. Configure and run distillation:
|
||||
|
||||
#### For DMD-only distillation:
|
||||
```bash
|
||||
bash examples/distill/Wan2.2-TI2V-5B-Diffusers/crush_smol/examples/distill/Wan2.2-TI2V-5B-Diffusers/crush_smol/
|
||||
```
|
||||
|
||||
@@ -77,7 +77,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.0
|
||||
--dit_precision "fp32"
|
||||
|
||||
@@ -69,7 +69,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--multi_phased_distill_schedule "4000-1"
|
||||
|
||||
@@ -103,7 +103,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--multi_phased_distill_schedule "4000-1"
|
||||
|
||||
@@ -101,7 +101,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--dit_precision "fp32"
|
||||
|
||||
@@ -101,7 +101,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--dit_precision "fp32"
|
||||
|
||||
@@ -69,7 +69,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--multi_phased_distill_schedule "4000-1"
|
||||
|
||||
@@ -103,7 +103,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--multi_phased_distill_schedule "4000-1"
|
||||
|
||||
@@ -69,7 +69,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--multi_phased_distill_schedule "4000-1"
|
||||
|
||||
@@ -69,7 +69,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--multi_phased_distill_schedule "4000-1"
|
||||
|
||||
@@ -100,7 +100,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--multi_phased_distill_schedule "4000-1"
|
||||
|
||||
@@ -69,7 +69,6 @@ optimizer_args=(
|
||||
# Miscellaneous arguments
|
||||
miscellaneous_args=(
|
||||
--inference_mode False
|
||||
--allow_tf32
|
||||
--checkpoints_total_limit 3
|
||||
--training_cfg_rate 0.1
|
||||
--multi_phased_distill_schedule "4000-1"
|
||||
|
||||
@@ -20,7 +20,7 @@ class SamplingParam:
|
||||
|
||||
# Text inputs
|
||||
prompt: str | list[str] | None = None
|
||||
negative_prompt: str | None = None
|
||||
negative_prompt: str = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"
|
||||
prompt_path: str | None = None
|
||||
output_path: str = "outputs/"
|
||||
output_video_name: str | None = None
|
||||
|
||||
@@ -1 +1 @@
|
||||
__version__ = "0.1.4"
|
||||
__version__ = "0.1.5"
|
||||
|
||||
+1
-1
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
|
||||
|
||||
[project]
|
||||
name = "fastvideo"
|
||||
version = "0.1.4"
|
||||
version = "0.1.5"
|
||||
description = "FastVideo"
|
||||
readme = "README.md"
|
||||
requires-python = ">=3.10"
|
||||
|
||||
@@ -35,7 +35,6 @@ torchrun --nnodes 1 --nproc_per_node $NUM_GPUS \
|
||||
--validation_sampling_steps "3" \
|
||||
--log_validation \
|
||||
--checkpoints_total_limit 3 \
|
||||
--allow_tf32 \
|
||||
--ema_start_step 0 \
|
||||
--training_cfg_rate 0.0 \
|
||||
--output_dir "outputs_dmd/wan_finetune" \
|
||||
|
||||
@@ -36,7 +36,6 @@ torchrun --nnodes 1 --nproc_per_node $NUM_GPUS \
|
||||
--validation_sampling_steps "3" \
|
||||
--log_validation \
|
||||
--checkpoints_total_limit 3 \
|
||||
--allow_tf32 \
|
||||
--ema_start_step 0 \
|
||||
--training_cfg_rate 0.0 \
|
||||
--output_dir "outputs_dmd/wan_finetune" \
|
||||
|
||||
@@ -35,7 +35,6 @@ torchrun --nnodes 1 --nproc_per_node $NUM_GPUS\
|
||||
--validation_sampling_steps "2,4,8" \
|
||||
--log_validation \
|
||||
--checkpoints_total_limit 3\
|
||||
--allow_tf32\
|
||||
--ema_start_step 0\
|
||||
--training_cfg_rate 0.0\
|
||||
--output_dir="$DATA_DIR/outputs/wan_finetune"\
|
||||
|
||||
@@ -41,7 +41,6 @@ torchrun --nnodes 1 --nproc_per_node $NUM_GPUS \
|
||||
--validation_sampling_steps "50" \
|
||||
--log_validation \
|
||||
--checkpoints_total_limit 3 \
|
||||
--allow_tf32 \
|
||||
--ema_start_step 0 \
|
||||
--training_cfg_rate 0.0 \
|
||||
--output_dir "$DATA_DIR/outputs/wan_finetune" \
|
||||
|
||||
@@ -11,9 +11,9 @@ fastvideo generate \
|
||||
--sp-size $num_gpus \
|
||||
--tp-size 1 \
|
||||
--num-gpus $num_gpus \
|
||||
--height 448 \
|
||||
--height 480 \
|
||||
--width 832 \
|
||||
--num-frames 61 \
|
||||
--num-frames 81 \
|
||||
--num-inference-steps 3 \
|
||||
--dit-cpu-offload False \
|
||||
--vae-cpu-offload False \
|
||||
@@ -42,7 +42,7 @@ fastvideo generate \
|
||||
--tp-size 1 \
|
||||
--num-gpus $num_gpus \
|
||||
--height 480 \
|
||||
--width 848 \
|
||||
--width 832 \
|
||||
--num-frames 81 \
|
||||
--num-inference-steps 3 \
|
||||
--fps 16 \
|
||||
@@ -78,4 +78,27 @@ fastvideo generate \
|
||||
--output-path outputs_video_dmd_14B_720P/ \
|
||||
--VSA-sparsity 0.9 \
|
||||
--dmd-denoising-steps "1000,757,522" \
|
||||
--enable_torch_compile
|
||||
|
||||
# 720P 5B
|
||||
num_gpus=1
|
||||
export FASTVIDEO_ATTENTION_BACKEND=FLASH_ATTN
|
||||
export MODEL_BASE=FastVideo/FastWan2.2-TI2V-5B-FullAttn-Diffusers
|
||||
# export MODEL_BASE=hunyuanvideo-community/HunyuanVideo
|
||||
# You can either use --prompt or --prompt-txt, but not both.
|
||||
fastvideo generate \
|
||||
--model-path $MODEL_BASE \
|
||||
--sp-size $num_gpus \
|
||||
--tp-size 1 \
|
||||
--num-gpus $num_gpus \
|
||||
--height 704 \
|
||||
--width 1280 \
|
||||
--num-frames 104 \
|
||||
--num-inference-steps 3 \
|
||||
--fps 24 \
|
||||
--prompt-txt assets/prompt.txt \
|
||||
--negative-prompt "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards" \
|
||||
--seed 1024 \
|
||||
--output-path outputs_video_dmd_5B_720P/ \
|
||||
--dmd-denoising-steps "1000,757,522" \
|
||||
--enable_torch_compile
|
||||
Reference in New Issue
Block a user