Inicio
Recursos
Mis obras
Models
Biblioteca de voces
Desarrolladores
Blog
Español
Precios
Iniciar sesión
Models/Video/MiniMax H3

MiniMax: MiniMax H3

  • Text to video
  • Image to video
API referenceUse in Generador de vídeo IA

Rich camera language, suited to narrative sequences and shots with obvious movement, and it accepts a last frame on its own.

Its resolution tiers do not line up with anything else: it takes 768p and 2k, and rejects 480p, 720p and 1080p outright.

Strengths

  • Always outputs an audio track, whatever `generateAudio` says
  • 2k output, plus 768p which nothing else accepts
  • A last frame can be supplied on its own

Limitations

  • Rejects 480p, 720p and 1080p
  • Ignores `seed`, so the same prompt cannot be reproduced
  • Ignores `ratio` entirely for image-to-video
  • No 4:5 or 5:4

Capabilities

CapabilityValue
Text to videoYes
Image to videoYes
First/last frame–No
Reference video–No
Native audio–No
Lip sync–No
Motion transfer–No

Specs and limits

ParameterValue
Aspect ratio16:9 · 4:3 · 1:1 · 3:4 · 9:16 · 21:9

How it is billed

按秒计费，分辨率越高越贵，生成前实时估价

No standard price for this model yet.

Alternatives

  • Seedance 2.0 ProWhen you need 1080p, or a reference video to drive the camera.

Call it

# Set LISTENHUB_API_KEY firstcurl -X POST "https://api.marswave.ai/openapi/v1/video-generation/generate" \-H "Authorization: Bearer $LISTENHUB_API_KEY" \-H "Content-Type: application/json" \-d '{"model": "MiniMax-H3","content": [{"type": "text","text": "A cinematic aerial shot of a quiet coastal city at sunrise"}]}'
  • API reference
  • SDK & CLI
  • Agent Skills

Similar models

  • Seedance 2.0 Pro
  • Seedance 2.0 Fast
  • PPixVerse
  • Wan 3.0
  • Wan 3.0 Prime