fal releases H3 Max: #1-ranked post-trained video model at 30x inference speed

isidentical · x · 2026-08-27

fal Research introduced H3 Max, a post-trained video model built on a MiniMax base. It ranks #1 in overall quality, prompt understanding, and aesthetics on both first-party and third-party independent benchmarks (Artificial Analysis, Design Arena), while generating a 5-second 720p video in under 3 seconds — 30x+ faster inference than peers. It's 50% off for launch week, positioning it as the fastest, cheapest top-quality video model.

Related event: fal's post-trained H3 Max tops video leaderboard with order-of-magnitude speedup(12 posts)→

Original post →

More from Multimodal

Multimodal channel →