FastH3-Omni Announced: Open Post-Trained Version of MiniMax's Realtime H3 Coming Oct 20
ostrisai · x · 2026-10-14
FastVideo, with Nuva Lab and Reactor, announced FastH3-Omni — the fastest, highest-quality open-source version of MiniMax's realtime video model H3, post-trained by @haoailab.
- Fully open-sourced October 20 with training code, supporting DGX Spark and RTX GPUs
- Exclusive preview via Reactor API until 10.20.2026
- Features: first/last frame conditioning, audio input with strong dialogue, up to 3 reference images, frontier-level performance
Related event: FastH3-Omni Set for October 20 Open-Source Release(2 posts)→
More from Multimodal
- World Labs-generated scenes run smoothly in AR via WebGPU: a parallax portal cube demo — yungcontent · 2026-10-14
- Noob Tip: Extract the Exact ComfyUI Workflow Back Out of Your Generated Images — zMarceeh · 2026-10-14
- Gemini Argon 4 High's Pelican Riding a Bicycle SVG — Thewolf421 · 2026-10-14
- FastH3-Omni lands Oct 20 with open weights, local RTX/Spark/MLX support — Vandy_simp · 2026-10-14
- AI short 'The Last Span' wins Astana fest award; video gen took just 1 of 6 weeks — lmoroney · 2026-10-14
- Dev builds an AI explainer-video bot with Grok, shares reusable template — XFreeze · 2026-10-14