QuantStack/Wan2.1_14B_VACE-GGUF
Same weights, converted vs Wan-AI/Wan2.1-VACE-14B. Same weights stored in another dtype (F32/BF16 → F16/F32).
Tensors changed: Attention 0/768, Norms 0/288, MLP 0/192, other 0/71, Embeddings 0/12
Same weights, converted vs Wan-AI/Wan2.1-VACE-14B. Same weights stored in another dtype (F32/BF16 → F16/F32).
Tensors changed: Attention 0/768, Norms 0/288, MLP 0/192, other 0/71, Embeddings 0/12