Instructions to use olka-fi/MiniMax-M2.7-MXFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use olka-fi/MiniMax-M2.7-MXFP4 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="olka-fi/MiniMax-M2.7-MXFP4", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("olka-fi/MiniMax-M2.7-MXFP4", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("olka-fi/MiniMax-M2.7-MXFP4", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use olka-fi/MiniMax-M2.7-MXFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "olka-fi/MiniMax-M2.7-MXFP4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "olka-fi/MiniMax-M2.7-MXFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/olka-fi/MiniMax-M2.7-MXFP4

SGLang

How to use olka-fi/MiniMax-M2.7-MXFP4 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 "olka-fi/MiniMax-M2.7-MXFP4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "olka-fi/MiniMax-M2.7-MXFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "olka-fi/MiniMax-M2.7-MXFP4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "olka-fi/MiniMax-M2.7-MXFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use olka-fi/MiniMax-M2.7-MXFP4 with Docker Model Runner:
```
docker model run hf.co/olka-fi/MiniMax-M2.7-MXFP4
```

request to put up 2.5 mxfp4

by Souhardya6x7 - opened Apr 13

Discussion

Souhardya6x7

Apr 13

please put up the minimax m2.5 mxfp4 version as well. I saw that it was there but then it suddenly got taken down

olka-fi

Owner Apr 13

Hi, m2.5 was broken in a sense - it required custom vllm weights unpacking that was never merged to mainline. I’ve deleted it because 2.7 beats it by all means.

But if you say that there’s need for m2.5 I’ll create proper quant for it.
Can expect by the end of this week

Souhardya6x7

Apr 13

i wanted to REAP it for AIMO-3 since m2.7 isnt allowed there... if its impossible by today or so, then atleast for me there's no need👍

olka-fi

Owner Apr 14

It takes ~16 hours only to upload model weights without testing, so there’s 0 chance to make it by today

Souhardya6x7 changed discussion status to closed Apr 14

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