H3 Max video model generates 5s clips in 3s, ranks #1 in quality
yuwen_lu_ · x · 2026-09-02
fal Research released H3 Max, a new post-trained video model. It claims the #1 spot for overall quality, prompt understanding, and aesthetics against leading models in independent evaluations. Performance-wise, it generates a 5-second 720p video in under 3 seconds, making it the fastest model tested, up to 29.8x faster than Seedance. The model is 50% off for the next week.
More from Multimodal
- Gemini upgrades video understanding with 88% reduction in token usage — haider1 · 2026-09-02
- Gemini agentic video understanding launches with a developer guide — osanseviero · 2026-09-02
- Gemini adds agentic video understanding, cutting token usage by 88% — osanseviero · 2026-09-02
- Fei-Fei Li's World Labs releases Atlas, a multimodal world model with pixel-perfect camera control — Chris_Armstrong · 2026-09-02
- AI Agents Collaborate on Filmmaking, Invideo Supports Series Universe Creation — LudovicCreator · 2026-09-02
- Gemini agentic video understanding: 88% fewer tokens, 66% lower cost, ~7% higher accuracy — _philschmid · 2026-09-02