tatsuyaaaaaaa/translategemma-4b-it-gguf

Same weights, converted (fewer layers) vs google/translategemma-4b-it. tatsuyaaaaaaa/translategemma-4b-it-gguf matches part of google/translategemma-4b-it (444 of its 883 tensors; the rest of google/translategemma-4b-it has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → BF16/F32).

Tensors changed: Embeddings 0/1, Norms 0/205, MLP 0/102, Attention 0/136

On the Hub · Downloads on Model Pulse