mlx-community/gemma-4-e4b-it-bf16

Same weights, repacked vs google/gemma-4-E4B-it. Same weights in different files (re-saved, re-sharded or new metadata), plus 665 new tensors.

Tensors changed: MLP 0/480, Norms 0/206, other 0/122, Attention 0/585, Embeddings 0/4

On the Hub · Downloads on Model Pulse