mradermacher/gemma-4-12B-it-untied-GGUF

Quantization vs allura-forge/gemma-4-12B-it-untied. mradermacher/gemma-4-12B-it-untied-GGUF is a GGUF quantization (Q8_0) of the same weights: 289 tensors match exactly and 330 within quantization error.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/289, MLP 0/144, Attention 0/184

On the Hub · Downloads on Model Pulse