CoolFace
Modelpublic

mlx-community/Fara1.5-4B-OptiQ-4bit

sourceHugging Facemitupdated 11d agoView on Hugging Face
2likes336downloads
4 commits on main
749fd4d11d ago

Update kv_config.json: re-measured with `optiq kv-cache` float32 sensitivity (the previous config was ranked on bfloat16 rounding noise). 8-bit layers [11] -> [7].

codelion
851901012d ago

Add kv_config.json: per-layer KV cache precision from `optiq kv-cache` (target 4.5 bits). Use with `optiq serve --kv-config kv_config.json`.

codelion
267d14f2mo ago

Add files using upload-large-folder tool

codelion
4e3222b2mo ago

initial commit

codelion