Instructions to use AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit 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("AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit")

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 AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit 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 "AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit"

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 AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit"

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

AX-MiniMax-M3-MLX-AXQ-2bit

Experimental AXQ 2-bit MLX pack of MiniMaxAI/MiniMax-M3 (BF16 source).

Not certified. Will not be certified in this revision. Layer-stack SSD expert paging is too slow for practical serving. Hobby / curiosity only. No AXQ 4-bit affine sibling. An MXFP4 SKU is a separate pack. Config num_mtp_modules is not packaged MTP, so this leaf has no -MTP.

Language path only. Vision stays BF16; vision generate is not claimed. Quality vs BF16 was not measured.

Item	Status
Convert + `ax_expert_stream.json`	Factory job on `df-macstudio-m2`
Convert git SHA	`2f4e4e49b82463ac9e146090020a9565c8583253`
License	Upstream MiniMax LICENSE copied into the pack
Stream	`--expert-stream required`

Downloads last month: 246

Safetensors

Model size

55B params

Tensor type

F32

U32

MLX

Hardware compatibility

2-bit

Model tree for AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit

Base model

MiniMaxAI/MiniMax-M3

Quantized

(58)

this model

Collection including AutomatosX/AX-MiniMax-M3-MLX-AXQ-2bit

MiniMax

Collection

MiniMax M3 MLX: experimental AXQ 2-bit and MXFP4 packs (not certified). • 2 items • Updated 4 days ago