---
license: other
license_name: minimax-m3
language:
- en
pipeline_tag: text-generation
tags:
- minimax
- minimax-m3
- moe
- jang
- mlx
- vmlx
- 2-bit
---

# MiniMax-M3 · REAP-40 · JANG_2L

> ## ⚠️ Requires vMLX ≥ v1.5.62
> Earlier vMLX builds contain a runtime cache bug that causes **repetition loops on long
> output**. This is an engine issue, not a weights issue — **update vMLX to v1.5.62 or
> later before running this model.** On v1.5.62+ generation is clean.

A space-efficient [MiniMax-M3](https://huggingface.co/MiniMaxAI) bundle for Apple Silicon:
**40 % REAP expert pruning** + **JANG_2L** mixed-precision quantization, ~**95 GB** — the
smaller, lower-RAM sibling of
[REAP-32](https://huggingface.co/JANGQ-AI/MiniMax-M3-REAP32-JANG_2L) (more pruning, less
resident memory). Runs comfortably on a single 128 GB Mac via [vMLX](https://github.com/) /
MLX.

## What this is
- **Base:** MiniMax-M3 (`model_type=minimax_m3_vl`) — MoE, GQA-4, MSA Lightning Indexer,
  vision tower.
- **Pruning:** REAP saliency pruning, **40 %** of routed experts removed (77 of 128 kept
  per MoE layer), highest-saliency experts retained.
- **Quantization (JANG_2L, affine, group size 64):**
  | tensor | bits |
  |---|---|
  | routed experts `gate_proj` / `up_proj` | 2 |
  | routed experts `down_proj` | **3** |
  | shared experts | 6 |
  | dense MLP (layers 0–2) | 6 |
  | attention q/k/v/o | 8 |
  | embeddings | 6 |
  | lm_head | 8 |
  | vision tower + projectors | 8 |
  | norms, router gate, MSA indexer | fp16 |

  `down_proj` is kept at 3-bit (the rest of the routed experts are 2-bit) for stable
  long-form coherency. The full per-module bit map is written into `config.json`
  (`quantization`) and applied automatically by the loader.

## Which one to use
- **REAP-40 (this repo, ~95 GB):** smaller, lower resident RAM, more headroom.
- **[REAP-32](https://huggingface.co/JANGQ-AI/MiniMax-M3-REAP32-JANG_2L) (~105 GB):** keeps
  more experts; higher quality ceiling, closer to the RAM limit on 128 GB machines.

## Usage
Load in vMLX (**v1.5.62+**); the engine autodetects `minimax_m3_vl` and applies the correct
settings (native MSA cache, paged cache off, per-module quant map). Sampling defaults ship
in `generation_config.json` (`temperature=1.0`, `top_p=0.95`).

## Attribution
- Quantization & packaging: **Jinho Jang** · eric@jangq.ai
- Base model © MiniMax, used under the MiniMax-M3 license.