Update README.md
Browse files
README.md
CHANGED
|
@@ -22,7 +22,10 @@ license: apache-2.0
|
|
| 22 |
- It achieves up to 2.1x speed up on a single **H100** GPU.
|
| 23 |
- Our model is trained on **77×768×1280** resolution, but it supports generating videos with any resolution.(quality may degrade).
|
| 24 |
- We set **VSA attention sparsity** to 0.9, and training runs for **1500 steps (~14 hours)**. You can tune this value from 0 to 0.9 to balance speed and performance for inference.
|
| 25 |
-
-
|
|
|
|
|
|
|
|
|
|
| 26 |
- Try it out on **FastVideo** — we support a wide range of GPUs from **H100** to **4090**
|
| 27 |
- We use [FastVideo 720P Synthetic Wan dataset](https://huggingface.co/datasets/FastVideo/Wan-Syn_77x768x1280_250k) for training.
|
| 28 |
|
|
|
|
| 22 |
- It achieves up to 2.1x speed up on a single **H100** GPU.
|
| 23 |
- Our model is trained on **77×768×1280** resolution, but it supports generating videos with any resolution.(quality may degrade).
|
| 24 |
- We set **VSA attention sparsity** to 0.9, and training runs for **1500 steps (~14 hours)**. You can tune this value from 0 to 0.9 to balance speed and performance for inference.
|
| 25 |
+
- Finetuning and inference scripts are available in the [FastVideo](https://github.com/hao-ai-lab/FastVideo) repository:
|
| 26 |
+
- [1 Node/GPU debugging finetuning script](https://github.com/hao-ai-lab/FastVideo/blob/main/scripts/finetune/finetune_v1_VSA.sh)
|
| 27 |
+
- [Slurm training example script](https://github.com/hao-ai-lab/FastVideo/blob/main/examples/training/finetune/Wan2.1-VSA/Wan-Syn-Data/T2V-14B-VSA.slurm)
|
| 28 |
+
- [Inference script](https://github.com/hao-ai-lab/FastVideo/blob/main/scripts/inference/v1_inference_wan_VSA.sh)
|
| 29 |
- Try it out on **FastVideo** — we support a wide range of GPUs from **H100** to **4090**
|
| 30 |
- We use [FastVideo 720P Synthetic Wan dataset](https://huggingface.co/datasets/FastVideo/Wan-Syn_77x768x1280_250k) for training.
|
| 31 |
|