MiniMax H3 Tops Video Editing Leaderboard, Open Weights Coming Soon

multimodalart · x · 2026-07-31

According to Artificial Analysis, the latest MiniMax video model, H3, ranks #1 in video editing and secures top 3 spots in both Text-to-Video and Image-to-Video.

The multimodal model accepts text, images, videos, and audio to generate 5-15s clips with native audio. MiniMax plans to release the model's weights under a community license, potentially making it the leading open-weight video model available.

Related event: MiniMax-H3 Tops Video Editing Leaderboard and Plans Open Source(6 posts)→

Original post →

More from Models

Models channel →