Instructions to use catalystsec/MiniMax-M2-3bit-DWQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use catalystsec/MiniMax-M2-3bit-DWQ 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("catalystsec/MiniMax-M2-3bit-DWQ")

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
LM Studio

Pi new

How to use catalystsec/MiniMax-M2-3bit-DWQ with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "catalystsec/MiniMax-M2-3bit-DWQ"

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": "catalystsec/MiniMax-M2-3bit-DWQ"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use catalystsec/MiniMax-M2-3bit-DWQ 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 "catalystsec/MiniMax-M2-3bit-DWQ"

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 catalystsec/MiniMax-M2-3bit-DWQ

Run Hermes

hermes

MLX LM

How to use catalystsec/MiniMax-M2-3bit-DWQ with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "catalystsec/MiniMax-M2-3bit-DWQ"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "catalystsec/MiniMax-M2-3bit-DWQ"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "catalystsec/MiniMax-M2-3bit-DWQ",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

catalystsec/MiniMax-M2-3bit-DWQ

This model was quantized to 3-bit using DWQ with mlx-lm version 0.28.4.

Parameter	Value
DWQ learning rate	3e-7
Batch size	1
Dataset	`allenai/tulu-3-sft-mixture`
Initial validation loss	0.146
Final validation loss	0.088
Relative KL reduction	≈40 %
Tokens processed	≈1.09 M

MMLU-PRO Benchmark

Model	Score
3-bit DWQ	66.1
3-bit	62.0

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("catalystsec/MiniMax-M2-3bit-DWQ")
prompt = "hello"

if tokenizer.chat_template is not None:
    prompt = tokenizer.apply_chat_template(
        [{"role": "user", "content": prompt}],
        add_generation_prompt=True,
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
print(response)