Reference-to-video on MiniMax H3 HighSpeed: hold people, places, motion and voices steady by supplying up to 9 images, 3 video clips and 3 audio clips, then direct the scene in one prompt. Clips run 4–15 seconds at 768p with sound.
Workbench
These examples were written for this endpoint. Replace the prompt and sample media URLs with your own accessible inputs, then check the current quote and model availability before submitting.
- Output5s
motel-room-reunion Reference to video from one image, 16:9 at 768p, five seconds, with rain and her quoted line.
- 768p
- 16:9
- 5s
- Output4s
motel-reference-4s 768p · 4s. One scene assembled from image, video and audio references.
- 768p
- 16:9
- 4s
- Output4s
boxing-ring-reference 768p · 4s. The boxer from the reference photo, now training in a ring.
- 768p
- 16:9
- 4s
This example starts at $0.072 per output second. The Playground updates the estimate when you change the settings.
