Instructions to use Goldkoron/MiniMax-M2.7 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Goldkoron/MiniMax-M2.7 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Goldkoron/MiniMax-M2.7",
	filename="MiniMax-M2.7-K_G_2.50.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Goldkoron/MiniMax-M2.7 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Goldkoron/MiniMax-M2.7
# Run inference directly in the terminal:
llama-cli -hf Goldkoron/MiniMax-M2.7

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Goldkoron/MiniMax-M2.7
# Run inference directly in the terminal:
llama-cli -hf Goldkoron/MiniMax-M2.7

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 Goldkoron/MiniMax-M2.7
# Run inference directly in the terminal:
./llama-cli -hf Goldkoron/MiniMax-M2.7

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 Goldkoron/MiniMax-M2.7
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Goldkoron/MiniMax-M2.7

Use Docker

docker model run hf.co/Goldkoron/MiniMax-M2.7

LM Studio
Jan
Ollama
How to use Goldkoron/MiniMax-M2.7 with Ollama:
```
ollama run hf.co/Goldkoron/MiniMax-M2.7
```

Unsloth Studio new

How to use Goldkoron/MiniMax-M2.7 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 Goldkoron/MiniMax-M2.7 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 Goldkoron/MiniMax-M2.7 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Goldkoron/MiniMax-M2.7 to start chatting

Pi new

How to use Goldkoron/MiniMax-M2.7 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Goldkoron/MiniMax-M2.7

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": "Goldkoron/MiniMax-M2.7"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Goldkoron/MiniMax-M2.7 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Goldkoron/MiniMax-M2.7

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 Goldkoron/MiniMax-M2.7

Run Hermes

hermes

Docker Model Runner
How to use Goldkoron/MiniMax-M2.7 with Docker Model Runner:
```
docker model run hf.co/Goldkoron/MiniMax-M2.7
```

Lemonade

How to use Goldkoron/MiniMax-M2.7 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Goldkoron/MiniMax-M2.7

Run and chat with the model

lemonade run user.MiniMax-M2.7-{{QUANT_TAG}}

List all available models

lemonade list

K_G_3.50 quant outperforming the K_G_4.00 quant I wonder if the 4 bit didn't quantize correctly?

by EclipseMist - opened 30 days ago

Discussion

EclipseMist

30 days ago

Not sure why but the 4 bit hallucinates way more often than the 3.5 bit, gives strangely worded responses and has worse code quality. Using Identical settings so I am wondering if the 4 bit quantized incorrectly.

Goldkoron

Owner 30 days ago

Hmm, there shouldn't be any problems with it, but this model is known for issues with Q4_K tensors in llama-cpp specifically on CUDA model loading so there might be more of those NaN issues than that one tensor Unsloth pointed out.

I ran all my internal testing on Vulkan so I didn't catch any of this though in my tests.

EclipseMist

30 days ago

Hmm, there shouldn't be any problems with it, but this model is known for issues with Q4_K tensors in llama-cpp specifically on CUDA model loading so there might be more of those NaN issues than that one tensor Unsloth pointed out.

I ran all my internal testing on Vulkan so I didn't catch any of this though in my tests.

im also running vulkan it doesn't make any logical sense to me because higher bpw typically also means higher quality but I use the model daily and 3.5 bit is defiantly better. Really have no clue why.

Goldkoron

Owner 30 days ago

im also running vulkan it doesn't make any logical sense to me because higher bpw typically also means higher quality but I use the model daily and 3.5 bit is defiantly better. Really have no clue why.

Interesting, yeah I am not sure why that would be. I overall didn't like this model for my personal use cases so I stopped using it after making the quants. Even at Q8 it was giving me rough results with lots of mistakes compared to Qwen3.5 397B

EclipseMist

30 days ago

im also running vulkan it doesn't make any logical sense to me because higher bpw typically also means higher quality but I use the model daily and 3.5 bit is defiantly better. Really have no clue why.

Interesting, yeah I am not sure why that would be. I overall didn't like this model for my personal use cases so I stopped using it after making the quants. Even at Q8 it was giving me rough results with lots of mistakes compared to Qwen3.5 397B

Huh interesting for me the 3.5 bit as been incredibly close to sonnet 4.5 and has been getting a lot of things sonnet 4.6 gets wrong correct. Although claude has been nerfed lately. Feels like its 90% of what sonnet 4.5 was.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment