H3 keeps getting faster. ⚡️
@sgl_project + VDN-H3 now push MiniMax H3 beyond 2× real-time denoising on 8× B200 - generating 14.4s of 768p video in 9.0s end-to-end after warmup, with no measured quality regression.
Open models compound through open ecosystems. 🚀
inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
open weights — A model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.
Why it matters
Faster open-source inferenceRunning a trained model to get answers — the phase where AI is actually used, as opposed to trained.Full definition → for video generation lowers compute cost per video and shows open weightsA model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.Full definition → and open inference stacks compounding gains together, relevant for teams deploying video-gen at scale.
