mradermacher/gemma-4-1B-0.8B-tiny-GGUF

Same weights, converted vs inference-optimization/gemma-4-1B-0.8B-tiny. Same weights stored in another dtype (BF16 → F16/F32), plus 31 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/55, MLP 0/18, other 0/6, Attention 0/23

On the Hub · Downloads on Model Pulse