Instructions to use kaitchup/MiniMax-M3-GGUF-MoQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use kaitchup/MiniMax-M3-GGUF-MoQ with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf kaitchup/MiniMax-M3-GGUF-MoQ
# Run inference directly in the terminal:
llama cli -hf kaitchup/MiniMax-M3-GGUF-MoQ

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf kaitchup/MiniMax-M3-GGUF-MoQ
# Run inference directly in the terminal:
llama cli -hf kaitchup/MiniMax-M3-GGUF-MoQ

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf kaitchup/MiniMax-M3-GGUF-MoQ
# Run inference directly in the terminal:
./llama-cli -hf kaitchup/MiniMax-M3-GGUF-MoQ

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf kaitchup/MiniMax-M3-GGUF-MoQ
# Run inference directly in the terminal:
./build/bin/llama-cli -hf kaitchup/MiniMax-M3-GGUF-MoQ

Use Docker

docker model run hf.co/kaitchup/MiniMax-M3-GGUF-MoQ

LM Studio
Jan
Ollama
How to use kaitchup/MiniMax-M3-GGUF-MoQ with Ollama:
```
ollama run hf.co/kaitchup/MiniMax-M3-GGUF-MoQ
```
Unsloth Desktop

How to use kaitchup/MiniMax-M3-GGUF-MoQ with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kaitchup/MiniMax-M3-GGUF-MoQ

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "kaitchup/MiniMax-M3-GGUF-MoQ"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use kaitchup/MiniMax-M3-GGUF-MoQ with Docker Model Runner:
```
docker model run hf.co/kaitchup/MiniMax-M3-GGUF-MoQ
```

Lemonade

How to use kaitchup/MiniMax-M3-GGUF-MoQ with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull kaitchup/MiniMax-M3-GGUF-MoQ

Run and chat with the model

lemonade run user.MiniMax-M3-GGUF-MoQ-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use kaitchup/MiniMax-M3-GGUF-MoQ with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kaitchup/MiniMax-M3-GGUF-MoQ

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 kaitchup/MiniMax-M3-GGUF-MoQ

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use kaitchup/MiniMax-M3-GGUF-MoQ with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kaitchup/MiniMax-M3-GGUF-MoQ

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 "kaitchup/MiniMax-M3-GGUF-MoQ" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

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

bnjmnmarie commited on Jun 30

Commit

a6875e7

verified ·

1 Parent(s): bb49e64

Update README.md

Browse files

Files changed (1) hide show

README.md +7 -1

README.md CHANGED Viewed

@@ -22,6 +22,12 @@ base_model:
 GGUF models made with the method ("Mixture of Quantizations") proposed by [Waleed Ahmad](https://huggingface.co/w-ahmad).
 I also used Unsloth [M3's imatrix](https://huggingface.co/unsloth/MiniMax-M3-GGUF) for calibration.
-Still under evaluation.
 * Compute Sponsorship: [Verda](https://verda.com/?utm_source=kaitchup.substack.com&utm_medium=referral&utm_content=m3quant). I used 2 B300s for quantization and evaluation.

 GGUF models made with the method ("Mixture of Quantizations") proposed by [Waleed Ahmad](https://huggingface.co/w-ahmad).
 I also used Unsloth [M3's imatrix](https://huggingface.co/unsloth/MiniMax-M3-GGUF) for calibration.
+More details and evaluation here:
+[MiniMax M3 GGUF Quantization: From 852 GB to ~150 GB Without Breaking Accuracy](https://kaitchup.substack.com/p/minimax-m3-gguf-quantization-from)
+![image](https://cdn-uploads.huggingface.co/production/uploads/64b93e6bd6c468ac7536607e/Wx2xnH4miXHQDZ8e4Avvw.png)
+Avoid using the MoQ-2.5.
 * Compute Sponsorship: [Verda](https://verda.com/?utm_source=kaitchup.substack.com&utm_medium=referral&utm_content=m3quant). I used 2 B300s for quantization and evaluation.