rariruluis/gemma-4-12B-it-qat-OptiQ-4bit-MTP

Same weights, repacked vs mlx-community/gemma-4-12B-it-qat-OptiQ-4bit. Same weights in different files (re-saved, re-sharded or new metadata), plus 48 new tensors.

Tensors changed: Embeddings 0/14, Norms 0/289, other 0/48, MLP 0/432, Attention 0/552

On the Hub · Downloads on Model Pulse