mlx-community/gemma-4-31B-it-qat-assistant-bf16

Same weights, repacked vs google/gemma-4-31B-it-qat-q4_0-unquantized-assistant. Same weights in different files (re-saved, re-sharded or new metadata).

Tensors changed: Embeddings 0/1, Norms 0/21, other 0/6, MLP 0/12, Attention 0/8

On the Hub · Downloads on Model Pulse