Soprano-1.1-80M-4bit

4-bit affine quantization of mlx-community/Soprano-1.1-80M-bf16, produced with FFAI 0.1.0's ffai convert (mlx-affine format, group_size=64).

Conversion

ffai convert mlx-community/Soprano-1.1-80M-bf16 --bits 4 \
    --upload-repo ekryski/Soprano-1.1-80M-4bit

See also

• FFAI — fast Apple Silicon LLM inference. Model.load("ekryski/Soprano-1.1-80M-4bit") runs this checkpoint end-to-end.
• FFAI quickstart
• FFAI quantization docs