Instructions to use MiniMaxAI/MiniMax-H3-Image-PE with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MiniMaxAI/MiniMax-H3-Image-PE with Transformers:

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

pipe = pipeline("text-generation", model="MiniMaxAI/MiniMax-H3-Image-PE")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("MiniMaxAI/MiniMax-H3-Image-PE")
model = AutoModelForMultimodalLM.from_pretrained("MiniMaxAI/MiniMax-H3-Image-PE", device_map="auto")
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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use MiniMaxAI/MiniMax-H3-Image-PE with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MiniMaxAI/MiniMax-H3-Image-PE"
# 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-H3-Image-PE",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MiniMaxAI/MiniMax-H3-Image-PE

SGLang

How to use MiniMaxAI/MiniMax-H3-Image-PE 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-H3-Image-PE" \
    --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-H3-Image-PE",
		"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 "MiniMaxAI/MiniMax-H3-Image-PE" \
        --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-H3-Image-PE",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MiniMaxAI/MiniMax-H3-Image-PE with Docker Model Runner:
```
docker model run hf.co/MiniMaxAI/MiniMax-H3-Image-PE
```

Add a Diffusers usage section (Modular Diffusers custom blocks)

by multimodalart HF Staff - opened about 12 hours ago

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

+62

-0

multimodalart

MiniMax org about 12 hours ago

Adds a short "Use with Diffusers" section: text-to-image and edit PE through the diffusers/minimax-h3-image-pe-blocks custom blocks, which wrap the released prompts and request settings and constrain the answer with XGrammar.

Notes for review:

The ~~19 GiB figure assumes the 4-bit loading PR on the blocks repo (https://huggingface.co/diffusers/minimax-h3-image-pe-blocks/discussions/1) is merged; without it the blocks dequantize to bf16 (~~52 GiB).
The blocks repo is private for now, so the section only works once it is public.
Tested through ModularPipeline on an RTX PRO 6000: text-to-image and edit, NVFP4 and bitsandbytes backends, all returning valid schema-constrained answers.

500kg

MiniMax org about 8 hours ago

•

edited about 8 hours ago

Thanks for the Diffusers section! Main has changed since this PR was opened (the license now points to the MiniMax-H3-Image repo, and "Edit PE" has been renamed to "Image editing PE"). Could you rebase onto the current main and:

use "Image editing PE" consistently na, and refer to the model as MiniMax-H3-Image-PE;
keep the section between "How to use Image editing PE" and "License";

We'll merge right after that.

multimodalart

MiniMax org about 7 hours ago

aight, made it compatible!

Add a Diffusers usage section (Modular Diffusers custom blocks)25653093

ryanlee-dev changed pull request status to merged about 6 hours ago

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