wnfldchen/gemma-4-12B-it-qat-w4a16-ct-heretic

Quantization vs wnfldchen/gemma-4-12B-it-qat-q4_0-unquantized-heretic. wnfldchen/gemma-4-12B-it-qat-w4a16-ct-heretic is a quantized version of the other (I32/BF16).

On the Hub · Downloads on Model Pulse