gemma-4-12B-it

4-bit affine quantization of google/gemma-4-12B-it, produced with FFAI 0.1.0's ffai convert (mlx-affine format, group_size=64).

Conversion

ffai convert google/gemma-4-12B-it --bits 4

See also

• FFAI — fast Apple Silicon LLM inference. Model.load("google/gemma-4-12B-it") runs this checkpoint end-to-end.
• FFAI quickstart
• FFAI quantization docs