This is a reupload of including a fp8 conversion for people who can't run the 1.3b model in 16 bit precision.
Source du modèle
Description de la source
The fp16 model, original from huggingface https://huggingface.co/alibaba-pai/Wan2.1-Fun-1.3B-InP/blob/main/diffusion_pytorch_model.safetensors . And a converted fp8e4m3fn version (make sure you download the one you want.)
Sources
1 sourceArtefacts du modèle
2 artefactswan21Fun13BInpI2vFp16_fp8e4m3fnAndFp16_1540447.safetensors
safetensors · FP8 · FULL · 1,46 GB · SHA-256 ce2a068cf976…d438 · Civitai
Téléchargerwan21Fun13BInpI2vFp16_fp8e4m3fnAndFp16_1540455.safetensors
safetensors · FP16 · FULL · 2,91 GB · SHA-256 cdab0deae983…55f9 · Civitai
TéléchargerExtraits de sources
2 extraitsSource context: 3997 downloads · Type Checkpoint · Base model Wan Video
Source context: 3589 downloads · Type Checkpoint · Base model Wan Video