Instructions to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF",
	filename="MiniMax-M2.7-REAP-139B-A10B-MXFP4_MOE.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

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 exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

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 exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

Ollama
How to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with Ollama:
```
ollama run hf.co/exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M
```

Unsloth Studio

How to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF to start chatting

How to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

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 exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with Docker Model Runner:
```
docker model run hf.co/exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M
```

Lemonade

How to use exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.MiniMax-M2.7-REAP-139B-A10B-GGUF-Q4_K_M

List all available models

lemonade list

Original Model Link : dervig/m51Lab-MiniMax-M2.7-REAP-139B-A10B

name: MiniMax-M2.7-REAP-139B-A10B-GGUF
base_model: MiniMaxAI/MiniMax-M2.7
license: other
license_name: modified-mit
license_link: https://hf.co/MiniMaxAI/MiniMax-M2.7/blob/main/LICENSE
pipeline_tag: text-generation
tasks: text-generation
language: en
library_name: llama.cpp
papers: https://arxiv.org/abs/2510.13999
tags:
- Cerebras
- MiniMaxAI
- M2.7
- REAP
- GGUF
- static quantization

Description

This is a 230 billion parameter MiniMax M2.7 model with 40% of its experts pruned with REAP (Router-weighted Expert Activation Pruning), then converted to GGUF with llama.cpp and static quantized.

Command sequence using source version of llama.cpp from source and /opt/homebrew/Cellar/llama.cpp/8940 (78433f606) llama-quantize:

hf download dervig/m51Lab-MiniMax-M2.7-REAP-139B-A10B
python -m convert_hf_to_gguf ~/.cache/huggingface/...
llama-quantize MiniMax-M2.5-REAP-139B-A10B-BF16.gguf MXFP4_MOE

Downloads last month: 401

GGUF

Model size

139B params

Architecture

minimax-m2

Hardware compatibility

3-bit

4-bit

5-bit

6-bit

8-bit

Model tree for exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF

Base model

MiniMaxAI/MiniMax-M2.7

Quantized

(109)

this model

Collection including exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF

MiniMaxAI

Collection

8 items • Updated 12 days ago

Paper for exdysa/MiniMax-M2.7-REAP-139B-A10B-GGUF

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

Paper • 2510.13999 • Published Oct 15, 2025 • 20