MiniMax H3 official demonstration excerpt
Model output documented by the source.
www.minimax.ioMiniMax's open-weight multimodal model for reference-guided audio-video creation.
No videos generated yet
Enter a prompt and click Generate to create your first AI video.
MiniMax H3 brings multiple creative tasks and modalities into one generation model. The official release includes open weights and reference-based video editing; deployment and commercial use remain subject to its model license.
Model output documented by the source.
www.minimax.ioThe references below identify the sources for this model’s capabilities. Confirm current access, request limits and pricing with your service provider before integration.
Where do MiniMax H3 specifications come from?Review generated results before use. Available modes and request limits depend on the service you choose; check the linked documentation before integration.