QuantStack/MAGREF_Wan2.1_I2V_14B-GGUF
Same weights, converted vs MAGREF-Video/MAGREF. Same weights stored in another dtype (F32 → F16/F32).
Tensors changed: Attention 0/800, Norms 0/280, MLP 0/160, other 0/53, Embeddings 0/10
Same weights, converted vs MAGREF-Video/MAGREF. Same weights stored in another dtype (F32 → F16/F32).
Tensors changed: Attention 0/800, Norms 0/280, MLP 0/160, other 0/53, Embeddings 0/10