Models/MiniMax/MiniMax Media API
Live5 featured endpoints

MiniMax API for Video, Speech, Music & Image Editing

Use selected MiniMax media models through Muapi for video generation, speech synthesis, music creation, and subject-reference image editing. A Muapi API key provides a consistent submit-and-poll workflow across these endpoints.This page covers media-generation APIs only, not general MiniMax text or chat models. Pricing and request parameters vary by endpoint; follow each model link for current details.

Hailuo 2.3 Pro

Selected MiniMax media API endpoints.1 featured endpoints
VideoLiveFeatured
Hailuo 2.3 Pro

Hailuo 2.3 Pro Text to Video

Generate cinematic video from a text prompt with the Hailuo 2.3 Pro endpoint.

Video generation
$0.63/request
Try Model

MiniMax H3 Max

Selected MiniMax media API endpoints.1 featured endpoints
VideoNew
MiniMax H3 Max

MiniMax H3 Max Text to Video

Create video from text with MiniMax H3 Max; consult the endpoint for current pricing and options.

Video generation
See current model pricing
Try Model

Speech

Selected MiniMax media API endpoints.1 featured endpoints
SpeechLive
Speech 2.6 HD

MiniMax Speech 2.6 HD

Generate high definition speech audio from text.

Text to audio
See current model pricing
Try Model

Music 3.0

Selected MiniMax media API endpoints.1 featured endpoints
MusicLive
Music 3.0

MiniMax Music 3.0

Generate vocal and instrumental music from a text brief.

Text to music
$0.20/song
Try Model

Image-01

Selected MiniMax media API endpoints.1 featured endpoints
Subject ReferenceLive
Image-01 Subject Reference

MiniMax Image-01 Subject Reference

Edit an image from a prompt while preserving the appearance of the supplied subject reference.

Subject-reference image editing
See current model pricing
Try Model

MiniMax API on Muapi

Access selected MiniMax media-generation endpoints through Muapi: Hailuo and H3 video, speech synthesis, music generation, and subject-reference image editing. Requests use Muapi endpoints and a Muapi API key.

This page covers media generation only; it does not provide MiniMax general text or chat API access. Model-specific parameters and current prices are shown on each linked endpoint.

What you can build

Generate video

Use Hailuo and H3 Max endpoints for prompt-to-video generation.

Create audio

Generate speech with MiniMax Speech or music with MiniMax Music.

Edit images with subject reference

Provide a source image and prompt to transform the scene while preserving the subject’s appearance.

MiniMax API endpoints

FamilyInputOutputStarting priceFull lineup
Hailuo 2.3 ProText promptVideo$0.63/request/minimax-hailuo
MiniMax H3 MaxText promptVideoSee current endpoint pricing/minimax-h3-max
MiniMax Speech 2.6TextSpeech audioSee current endpoint pricing/text-to-speech
MiniMax Music 3.0Text briefMusic$0.20/song/minimax-music-3.0
MiniMax Image-01 Subject ReferenceSource image and promptEdited imageSee current endpoint pricing/ai-character-consistency-api

Prices and availability can change. Check each model endpoint for current details before submitting.

How to call the MiniMax API

Use a Muapi key to submit a model-specific request. Save the returned request ID and poll the shared result endpoint. The subject-reference image endpoint requires both a prompt and source image URL.

Submit a Hailuo video request

curl -X POST https://api.muapi.ai/api/v1/minimax-hailuo-2.3-pro-t2v \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A cinematic train crossing a bridge at blue hour"}'

# Response: {"request_id":"REQUEST_ID"}

Send a prompt and duration to Hailuo 2.3 Pro.

Submit a MiniMax Music request

curl -X POST https://api.muapi.ai/api/v1/minimax-music-3.0 \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Warm acoustic folk song with guitar and soft harmonies","lyrics":"We follow the morning road\nWith the first light shining ahead"}'

# Response: {"request_id":"REQUEST_ID"}

Generate a song from a text prompt.

Submit a MiniMax speech request

curl -X POST https://api.muapi.ai/api/v1/minimax-speech-2.6-hd \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Welcome to Muapi. Your workspace is ready.","voice_id":"Friendly_Person","speed":1,"emotion":"happy","format":"mp3"}'

# Response: {"request_id":"REQUEST_ID"}

Generate speech audio from text.

Submit a subject-reference image edit

curl -X POST https://api.muapi.ai/api/v1/minimax-01-subject-reference \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Place the subject in a sunlit garden while preserving their face and clothing","image_url":"https://example.com/portrait.jpg","aspect_ratio":"1:1","num_images":1}'

# Response: {"request_id":"REQUEST_ID"}

Send a prompt and source image to preserve a subject while changing the scene.

Poll for the result

curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result \
  -H "x-api-key: YOUR_API_KEY"

Check completion and read the generated output.

MiniMax API FAQ

Which MiniMax models are included?

This page features selected MiniMax media endpoints for video, speech, music, and image generation. The linked model pages show their full endpoint options.

Does this include MiniMax chat or text models?

No. This page is scoped to media generation endpoints and does not claim general MiniMax text or chat API access.

Do MiniMax endpoints share one request format?

No. Each model endpoint has its own request fields. Use the examples and model schemas linked from this page.

How do I authenticate?

Create a Muapi API key and send it in the x-api-key header to the selected model endpoint.

Build with MiniMax media models

Use selected MiniMax media models through Muapi for video generation, speech synthesis, music creation, and subject-reference image editing. A Muapi API key provides a consistent submit-and-poll workflow across these endpoints.