metadata
license: apache-2.0
base_model: allenai/OLMo-2-0425-1B-Instruct
language:
- en
tags:
- mlx
- ffai
- quantized
- 4bit
- affine
OLMo-2-0425-1B-Instruct-4bit
4-bit affine quantization of allenai/OLMo-2-0425-1B-Instruct, produced with FFAI 0.1.0's ffai convert (mlx-affine format, group_size=64).
Conversion
ffai convert allenai/OLMo-2-0425-1B-Instruct --bits 4 \
--upload-repo ekryski/OLMo-2-0425-1B-Instruct-4bit
See also
- FFAI — fast Apple Silicon LLM inference.
Model.load("ekryski/OLMo-2-0425-1B-Instruct-4bit")runs this checkpoint end-to-end. - FFAI quickstart
- FFAI quantization docs