MiniMax H3 generates 2K multimodal videos with multi-reference and native stereo audio.
|
Founded year:
|
2026
|
|
Country:
|
United States of America
|
|
Funding rounds:
|
Not set
|
|
Total funding amount:
|
Not set
|
Description
MiniMax H3 is MiniMax’s third-generation universal multimodal video generation model. It supports text-to-video, image-to-video, video-to-video transfer and AI video editing. It enables simultaneous reference of images, videos and audios to maintain consistent characters, motions and styles. Delivering up to 2K, 24fps videos with native stereo audio, it supports various aspect ratios. Unified architecture simplifies multimodal creation for short videos, ads and virtual human content.