MiniMax H3 LoRA reference-to-video combines up to nine images, three video clips and three audio clips with up to three of your own LoRAs. Refer to each input by number to get a 3 to 15 second shot at 480p to 1080p with stereo 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.
- Output7s
character-lora-named-refs Two stills, one for the woman and one for the room, with a realism LoRA at 0.8 on top. The references pin who and where; the prompt writes what happens.
- 42
- 540p
- 9:16
- Output7s
character-lora-plus-references Two stills, one for the woman and one for the room, with a realism LoRA at 0.8 on top. The references pin who and where; the prompt writes what happens.
- 42
- 540p
- 9:16
- Output5s
copy-the-spin-from-a-clip Picture 1 supplies the woman, Video 1 only the floor spin. Our first attempt without that sentence kept the dancer from the clip; saying which input brings the person and which brings the motion fixed it.
- 42
- 540p
- 16:9
This example starts at $0.06 per output second. The Playground updates the estimate when you change the settings.