WaveCut/Wan2.1-T2V-1.3B-Diffusers-OrbitQuant-W4A6

Same weights, converted vs Wan-AI/Wan2.1-T2V-1.3B-Diffusers. Same weights stored in another dtype (F32 → U8/BF16), plus 600 new tensors.

Tensors changed: Norms 0/180, Attention 0/240, MLP 0/60, other 0/33, Embeddings 0/12

On the Hub · Downloads on Model Pulse