Reference-led direction
communicate subjects and settings with selected images.
Video generation
Develop a reference-led shot with a wide resolution range.
Loading model settings…
Your prompt, settings, and references stay together in a new Canvas.
01 / Meet the model
MiniMax H3 supports prompt, frame, and visual-reference video creation. Its resolution ladder includes intermediate choices as well as a high-resolution output, while generated audio stays enabled. Use it for planning a scene whose composition is already partly defined by images.
02 / Choose it for
communicate subjects and settings with selected images.
set opening and closing images for a planned transition.
compare a working draft with a larger delivery candidate.
03 / Before you create
Model settings. Availability, input mode, and references can narrow the choices offered in Canvas.
04 / A closer look
Adaptive framing uses the supplied imagery to guide the output shape. In the frame workflow, WeGen selects adaptive framing; choose compatible reference compositions and inspect the resulting crop.
Text prompt · Reference images · Start frame · End frame. The input controls above show the options available for your selected mode.
Adaptive · 1:1 · 16:9 · 21:9 · 3:4 · 4:3 · 9:16. 480p · 768p · 1440p · 4K. The composer on this page accepts image references and frames; video/audio reference uploads are not exposed here. Frame mode follows the starting image's aspect ratio. Adaptive framing is offered with references.
Start with the format you intend to publish, then choose the available output settings. Changing inputs can change the choices. Review the current Gem estimate before Generate; availability and cost can vary.