Instructions to use Kijai/WanVideo_comfy_fp8_scaled with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusion Single File
How to use Kijai/WanVideo_comfy_fp8_scaled with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
|
Download Wan22Animate/README.md from Kijai/WanVideo_comfy_fp8_scaled: direct link, hf CLI and curl.
- Browser
- Download file 572 Bytes
-
https://huggingface.co/Kijai/WanVideo_comfy_fp8_scaled/resolve/main/Wan22Animate/README.md
- Command line
-
hf download hf://Kijai/WanVideo_comfy_fp8_scaled/Wan22Animate/README.md
-
curl -L -o README.md https://huggingface.co/Kijai/WanVideo_comfy_fp8_scaled/resolve/main/Wan22Animate/README.md
572 Bytes
WanAnimate fp8_scaled models
v2 bugfix update for native workflows
The initial uploads has the face encoder layers in bf16, as back then I was unsure if they can be quantized.
When running these models in the WanVideoWrapper those layers are used in the base_precision, but when running these in native ComfyUI, they get downscaled to normal fp8, which seems to introduce pretty bad noise in the outputs in a grid pattern.
The v2 update addresses this by using fp8 scaled on the face encoder layers as well. For the wrapper you can keep using the initial models.