gaianet/gemma-3n-E2B-it-GGUF

Same weights, converted (fewer layers) vs google/gemma-3n-E2B-it. gaianet/gemma-3n-E2B-it-GGUF matches part of google/gemma-3n-E2B-it (694 of its 1556 tensors; the rest of google/gemma-3n-E2B-it has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32), plus 32 new tensors.

Tensors changed: Embeddings 0/1, other 0/211, Norms 0/272, MLP 0/90, Attention 0/120

On the Hub · Downloads on Model Pulse