ManniX-ITA/gemma-4-A4B-98e-v6-coder-it-GGUF

Same weights, converted vs ManniX-ITA/gemma-4-A4B-98e-v6-coder-it. Same weights stored in another dtype (BF16 → F16/F32), plus 151 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/271, MLP 0/90, other 0/30, Attention 0/115

On the Hub · Downloads on Model Pulse