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

Libraries

How to use MiniMaxAI/MiniMax-M3 with Transformers:

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

pipe = pipeline("image-text-to-text", model="MiniMaxAI/MiniMax-M3", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("MiniMaxAI/MiniMax-M3", trust_remote_code=True)
model = AutoModelForMultimodalLM.from_pretrained("MiniMaxAI/MiniMax-M3", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use MiniMaxAI/MiniMax-M3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MiniMaxAI/MiniMax-M3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MiniMaxAI/MiniMax-M3",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/MiniMaxAI/MiniMax-M3

SGLang

How to use MiniMaxAI/MiniMax-M3 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-M3" \
    --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": "MiniMaxAI/MiniMax-M3",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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-M3" \
        --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": "MiniMaxAI/MiniMax-M3",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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

Anyone get this working on 2x rtx 6000 pro?

#10

by Quadrapole - opened 4 days ago

Discussion

Quadrapole

4 days ago

Would love to see how you got it running and if using q3

Or how it runs at q3

Thanks

llmfan46

about 17 hours ago

•

edited about 16 hours ago

2x RTX PRO 6000 = 192 GB

This model = 854 GB

So nope, not possible.

About Q3, I assume you mean Q3 GGUFs (Q3_K_L, Q3_K_M, Q3_K_S), the issue is that GGUF support is not quite ready yet, it's still a work in progress, see here:

https://github.com/ggml-org/llama.cpp/pull/24523

Now you might then be wondering how Unsloth posted GGUFs then, here:

https://huggingface.co/unsloth/MiniMax-M3-GGUF

That's because Unsloth released EXPERIMENTAL GGUFs with the partial preliminary support available from PR 24523, but as it's not quite ready yet as it's only preliminaries and so there are many things don't work yet, things that are not supported and things that need fixes, like the vision projector, if you look at the Unsloth GGUFs, there is no vision projector in there, another thing is that sparse attention (MiniMaxM3SparseForConditionalGeneration) does not work either and it falls back to dense attention instead (Unsloth says it on their model card here: Note: MiniMax Sparse Attention is not supported yet, so inference falls back to dense attention.) So basically the full multimodal model doesn't have GGUF support yet, it's text-only with no support for sparse attention and tool calling need fixes.

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