MiniMax H3 API: Python SDK for AI Video Generation

Python SDK for the MiniMax H3 API on Muapi. Generate video from text, animate a still image, or guide a new video with image, video, and optional audio references using one API key.

▶ Watch: How to Access MiniMax Hailuo H3 API (Step-by-Step Guide): Native 2K Video Generation with Sound

Related Projects

• MiniMax H3 text-to-video playground — Try the primary SDK workflow in the browser before integrating it.
• awesome-minimax-h3-prompts — Prompt gallery with runnable MuAPI examples and creator-ready video references.
• Open-Generative-AI — open-source studio and model hub for generative image and video workflows.
• awesome-ai-video-models — compare leading AI video models, API access, and pricing.
• Seedance-2-API — Python SDK for ByteDance Seedance text-to-video and image-to-video workflows.
• Veo-4-API — Python SDK for Google Veo video-generation workflows.
• Flux-3-Dev-API — unified image and video SDK for the FLUX 3 family.
• Generative-Media-Skills — agent-ready skills for building generative-media pipelines.
• muapi-cli — CLI and MCP access to Muapi generation tasks.

Features

• Text-to-video generation
• Image-to-video animation
• Reference-to-video generation with multimodal inputs
• First-and-last-frame guidance through image-to-video
• Submit, poll, and webhook-ready asynchronous workflow
• Simple Python client built on requests

Installation

pip install minimax-h3-api

Or install from source:

git clone https://github.com/Anil-matcha/MiniMax-H3-API.git
cd MiniMax-H3-API
pip install -e .

Set your Muapi API key:

export MUAPI_API_KEY=your_muapi_api_key

Quick start

from minimax_h3_api import MiniMaxH3API

api = MiniMaxH3API()

task = api.text_to_video(
    "A cinematic tracking shot of a silver sports car driving through a rain-soaked city at night",
    aspect_ratio="16:9",
    resolution="2k",
    duration=5,
)

result = api.wait_for_completion(task["request_id"])
print(result["outputs"][0])

Workflows

Text to video

task = api.text_to_video(
    prompt="A lone astronaut walking through a field of bioluminescent flowers, slow camera dolly in"
)

Image to video

task = api.image_to_video(
    prompt="The camera slowly pushes in as a warm breeze moves through the subject's hair",
    image_url="https://example.com/source-image.jpg",
    last_image_url="https://example.com/final-frame.jpg",
    duration=5,
)

Reference to video

task = api.reference_to_video(
    prompt="Use the references to create a cinematic product reveal with a slow camera orbit",
    reference_images=["https://example.com/product.jpg"],
    reference_videos=["https://example.com/motion-reference.mp4"],
    reference_audios=["https://example.com/soundtrack.mp3"],
    aspect_ratio="16:9",
    resolution="2k",
    duration=5,
)

At least one reference_images or reference_videos URL is required. Audio references are optional and cannot be used alone.

First and last frame compatibility alias

Existing integrations can continue using first_last_frame(). It now maps to the current image-to-video endpoint with image_url and last_image_url; new code should call image_to_video() directly.

API endpoints

Workflow	Endpoint
Text to Video	POST /api/v1/minimax-h3-text-to-video
Image to Video	POST /api/v1/minimax-h3-image-to-video
Reference to Video	POST /api/v1/minimax-h3-reference-to-video
Poll task	GET /api/v1/predictions/{request_id}/result

See the full MiniMax H3 API guide and Muapi documentation.

License

MIT