MiniMax H3 Max Image to Video

Image inputs

Field	Purpose
image_url	Required starting frame
end_image_url	Optional ending keyframe
prompt	Motion, camera and audio direction
duration	5 to 15 seconds
resolution	480P or 768P

Example request

Example

curl -X POST "https://api.oneinfer.ai/v1/ula/generate-video" \
  -H "Authorization: Bearer $ONEINFER_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Minimax/minimax-h3-max",
    "prompt": "Camera slowly pulls back while rain ripples across the street",
    "image_url": "https://example.com/opening-frame.jpg",
    "resolution": "768P",
    "duration": 5
  }'

Ready to test the workflow?

Create account & add credits

Aspect ratio comes from the image

Unlike text-to-video, image-to-video derives the output aspect ratio from the uploaded starting frame. Crop the source deliberately before submitting it.

Frequently asked questions

Can H3 Max use a first and last frame?

Yes. Send image_url for the opening frame and end_image_url for the ending keyframe.

What model ID should I use on OneInfer?

Use "Minimax/minimax-h3-max" exactly, including capitalization. The MiniMax provider model name is MiniMax-H3-Max, but that is not the OneInfer model ID.

How long can a MiniMax H3 Max video be?

H3 Max generates clips from 5 to 15 seconds at 24 fps.

Does MiniMax H3 Max generate audio?

Yes. It predicts synchronized audio alongside the video in the same generation.

Put MiniMax H3 Max to work

Fund a controlled evaluation, send a reference frame or document, and measure quality and cost on your own workload.

Create account & add credits Run this with MiniMax H3 Max