README.md

metadata

library_name: mlx
tags:
  - mlx
  - oq
  - quantized

MiniMax-M3-oQ3

This model was quantized using oQ (oMLX v0.4.5.dev1) mixed-precision quantization.

Quantization details

• Model type: minimax_m3_vl
• Bits: 3
• Group size: 64
• Format: MLX safetensors