Instructions to use MiniMaxAI/MiniMax-M2.7 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MiniMaxAI/MiniMax-M2.7 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="MiniMaxAI/MiniMax-M2.7")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("MiniMaxAI/MiniMax-M2.7", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use MiniMaxAI/MiniMax-M2.7 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "MiniMaxAI/MiniMax-M2.7" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MiniMaxAI/MiniMax-M2.7", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/MiniMaxAI/MiniMax-M2.7
- SGLang
How to use MiniMaxAI/MiniMax-M2.7 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "MiniMaxAI/MiniMax-M2.7" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MiniMaxAI/MiniMax-M2.7", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "MiniMaxAI/MiniMax-M2.7" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MiniMaxAI/MiniMax-M2.7", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use MiniMaxAI/MiniMax-M2.7 with Docker Model Runner:
docker model run hf.co/MiniMaxAI/MiniMax-M2.7
(Un)officially support 1x96GB?
Can't wait to try this, thanks.
Worth noting: M2.5 actually runs pretty well at UD-IQ2_M, in 96GB. A bit limited in context, but it works. That brings it within reach of non-commercial enthusiasts with 4x3090, or 3x32GB.
I notice that your recommended configs are for 4x96GB and 8x96GB, so you have the right hardware, and could "unofficially support" 1x96GB too :)
This is really just me begging you to keep (at least the lower end of) your models at 200-220B, even if you add multimodality in future. It's a great level of complexity, and just within reach for a lot of us, but I suppose it could help you get more attention too, if you unofficially support systems that more people have.