DomoAI Launches Seedance 2.5 and MiniMax H3 in Omni Reference for Long-Form Character Storytelling and Music Video Creation
DomoAI today launched Seedance 2.5 and MiniMax H3 in Omni Reference, giving creators two ways to build longer projects around the same character: scene-by-scene storytelling and music-led performance.
Seedance 2.5 is aimed at a problem familiar to anyone making an animated series or motion comic. A character may look right in one shot, then change face, clothing, or proportions in the next. Omni Reference lets creators carry the same visual references across scenes, using up to 30 images, 10 video clips, 10 audio files, and first and last frames to guide the result.
Japanese creator JPN GIRL recently used the workflow to make an 80-second AI short film in DomoAI. The film was assembled from multiple generated shots, showing how a creator can move beyond isolated clips and build a continuous scene with recurring characters.
Read More – What Problems Does Collaborative AI Video Solve for Marketing Teams?
MiniMax H3 takes the same idea into music video production. Creators can combine a song with up to nine image references, three videos, and three audio files. The audio helps guide timing, rhythm, vocal performance, and movement, while the visual references establish who the character is and how the shot should feel.
MiniMax H3 generates individual clips between four and 15 seconds in 768P or up to 2K. Those clips can then be arranged around verses, choruses, and other cut points rather than treating a full music video as a single generation.
Write to us [wasim.a@demandmediaagency.com] to learn more about our exclusive editorial packages and programmes.
