AnalysisVisual AIAugust 9, 2026

MiniMax H3 video-to-video demo shows model replacing objects in footage

A Reddit user demonstrates MiniMax H3's video-to-video capability, using a prompt to replace a giant cat with a dog in a city scene. The model follows detailed instructions to alter the video content.

1 source

Daily brief

Get tomorrow's AI brief in your inbox

More stories today

Open the live feed