Instructions to use JANGQ-AI/MiniMax-M3-REAP22-Coder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use JANGQ-AI/MiniMax-M3-REAP22-Coder with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("JANGQ-AI/MiniMax-M3-REAP22-Coder")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use JANGQ-AI/MiniMax-M3-REAP22-Coder with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "JANGQ-AI/MiniMax-M3-REAP22-Coder"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "JANGQ-AI/MiniMax-M3-REAP22-Coder"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use JANGQ-AI/MiniMax-M3-REAP22-Coder with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "JANGQ-AI/MiniMax-M3-REAP22-Coder"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default JANGQ-AI/MiniMax-M3-REAP22-Coder

Run Hermes

hermes

MLX LM

How to use JANGQ-AI/MiniMax-M3-REAP22-Coder with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "JANGQ-AI/MiniMax-M3-REAP22-Coder"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "JANGQ-AI/MiniMax-M3-REAP22-Coder"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "JANGQ-AI/MiniMax-M3-REAP22-Coder",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

MiniMax-M3-REAP22-Coder

A JANG-quantized MiniMax-M3 — coding/agentic + multimodal — for the vMLX engine (Apple Silicon / MLX).

⚠️ Requires vMLX engine v1.5.67 or newer. This is a JANG-format model (JANG affine-mixed + AWQ quantization, REAP expert pruning, and the MiniMax-M3 MSA / Lightning-Indexer runtime). It will NOT load with transformers, vLLM, or generic MLX loaders — it needs vMLX's JANG loader + the M3 runtime. Coder support lands in vMLX ≥ 1.5.67.

What is a JANG model?

JANG is vMLX's quantization + packing format: mixed-precision affine quantization (per-projection bit widths) + AWQ activation-aware scaling + REAP expert pruning, described by a jang_config.json. Weights stay quantized in GPU memory and are loaded by vMLX's JANG loader. Because the format and the MiniMax-M3 runtime (MSA dual-cache, Lightning Indexer, partial RoPE, vision tower) are vMLX-specific, these models run only on vMLX ≥ 1.5.67.

Run it

Install/update vMLX 1.5.67+ — https://mlx.studio (or pip install -U vmlx).
App: Server → New Session → pick/download this model → Start → chat.
CLI: vmlx-engine serve JANGQ-AI/MiniMax-M3-REAP22-Coder --reasoning-parser minimax_m3 --tool-call-parser minimax_m3

Highlights

Coding: HumanEval pass@1 = 100% (81/81 on a scrambled half of HumanEval, first-sample) — pass@5 = 1.000.
Arithmetic/reasoning recovered vs the base REAP quant (with reasoning enabled): ~7/7 on a 7-task probe.
Multimodal (vision) kept. ~107 GB on disk.

Build

Base: MiniMaxAI/MiniMax-M3 (60 layers, MoE, MSA Lightning Indexer, GQA, partial RoPE).
REAP pruning: keep 100/128 routed experts per MoE layer (22% pruned), saliency-scored.
JANG affine quant (group_size 64): routed gate/up = 2-bit + AWQ pre-scaling, down = 2-bit; shared experts 6-bit; attention 8-bit; embeddings 6-bit; lm_head 8-bit; Lightning Indexer + norms FP16; vision 8-bit.
"Floor" expert recipe: protect the proven coding experts (coding saliency) + add top math experts, so coding stays intact while math improves.
Calibration: Vera (agentic-coder) dominant + GSM8K (math reasoning).

Attribution

Base model: MiniMaxAI/MiniMax-M3
Expert pruning: REAP (Cerebras, ICLR 2026, arXiv:2510.13999)
Vera agentic-coder calibration dataset + evaluation/testing: @hornsman1 (hornsan1 on GitHub)
Additional math-reasoning calibration: GSM8K
Quantization & runtime: JANG / vMLX

Credits

Vera dataset & model testing: @hornsman1 (hornsan1 on GitHub)

Downloads last month: -

Safetensors

Model size

34B params

Tensor type

U32

F16

MLX

Hardware compatibility

Quantized

Model tree for JANGQ-AI/MiniMax-M3-REAP22-Coder

Base model

MiniMaxAI/MiniMax-M3

Finetuned

(7)

this model

Collection including JANGQ-AI/MiniMax-M3-REAP22-Coder

MiniMax M3 for 128gb

Collection

MiniMax m3 for Mac’s with 128gb RAM. • 4 items • Updated about 20 hours ago

Paper for JANGQ-AI/MiniMax-M3-REAP22-Coder

REAP the Experts: Why Pruning Prevails for One-Shot MoE compression

Paper • 2510.13999 • Published Oct 15, 2025 • 20