This is the FancyVideo vae_3d_16_frames model (https://huggingface.co/qihoo360/FancyVideo), but with its keys pruned and modified to be readily loadable by ComfyUI (or other AnimateDiff-compatible backends).
There is a FancyVideo key added to make the model easily distinguishable for either AnimateDiff-like models. It's value can be ignored, but it is a tensor of size(0) with value of 1.0 to distinguish it as version 1.0.
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no library tag.