Video Delta Net speeds up Minimax-H3 video generation 75-90x, faster than playback
xiuyu_l · x · 2026-09-03
Open-source video generation gets a speed breakthrough: Video Delta Net (VDN) uses hybrid attention for near-lossless live text-to-video, accelerating Minimax-H3 by 75–90x — generating 14 seconds of 768p video in just 11 seconds on 8× NVIDIA B200 GPUs, faster than playback speed with negligible quality loss.
The researchers (working with Haocheng Xi of UC Berkeley and team) argue video generation has long faced a speed–quality trade-off, and VDN shows you don't have to choose. Checkpoints, training/inference code, and a technical blog have been released.
Related event: VideoDeltaNet Speeds Open Video Generation Up to 90x(4 posts)→
More from Multimodal
- Runway's Solaris: an interface world model that renders apps pixel by pixel — umpherj · 2026-09-03
- VideoDeltaNet open-sources hybrid attention that speeds up MiniMax H3 video generation up to 90x — realmrfakename · 2026-09-03
- jjk-explain turns any concept into a Jujutsu Kaisen-style explainer video with one Claude Code command — teortaxesTex · 2026-09-03
- New ComfyUI node adds WYSIWYG video cropping with 8 fixed ratios on the Load Video preview — MayaProphecy · 2026-09-03
- WAN 3 Tops AI Video Editing Chart, Ranked #1 With Audio at 1189 Elo — koltregaskes · 2026-09-03
- MiniMax H3 open-weight model generates character-drawing timelapses with cursor UI — WolframRvnwlf · 2026-09-03