Open-weight Hailuo H3 handles text, image, audio and video in one model
- Source
- x.com
- Date
The new open-weights @Hailuo_AI H3 is live on OpenRouter thanks to @MiniMax_AI! H3 is a lightweight video model for text, image, audio, & video input, with native audio-visual output. Built for instruction-led edits, text/brand rendering, and video-2-video motion transfer.
Access all the advanced capabilities of H3 through OpenRouter's Video API, including first and last frame conditioning, image, video, and audio reference URLs (up to 9 images, 3 videos, 3 audio, 12 files total). Read more about passthrough parameters: https://t.co/8blC0FF2Ow
Context
MiniMax's Hailuo H3 gives video creators several ways to constrain a generated result. OpenRouter's announcement describes first- and last-frame conditioning, which supplies images for the beginning and end of a clip, alongside image, video and audio references. Those references have separate limits of nine images, three videos and three audio files, with a combined maximum of twelve files. The combined cap matters: the individual allowances cannot all be filled at once. The announcement describes the model as built for instruction-led editing and motion transfer, with audio and video produced together.
Checking sign-in…
Loading comments…