MiniMax H3

Create 4–15 second videos in 768P or 2K from text, one image, first and last frames, or Reference-to-Video inputs.

Short social videos Animating still images and references

Text-to-video, single-image I2V, first/last frames, and Reference-to-Video Reference inputs: up to 9 images, 3 videos, and 3 audio files 768P and 2K output 4–15 second clips

Use first and last frames for controlled transitions Add only the references needed to preserve the subject and motion

A cinematic drone shot over a misty forest at sunrise, slow camera movement

Use first and last frames for controlled transitions A cinematic drone shot over a misty forest at sunrise, slow camera movement