xv0y5ncu/gemma-4-E4B-it-GLQ-trellis-3inst-4bpw

Same weights, repacked vs google/gemma-4-E4B-it. Same weights in different files (re-saved, re-sharded or new metadata), plus 1374 new tensors.

Tensors changed: MLP 0/480, Norms 0/502, other 0/177, Attention 0/623, Embeddings 0/5

On the Hub · Downloads on Model Pulse