fal launches FLUX 3 Max: tied #1 in image-to-video, 10s clips in ~12 seconds, 75% off launch
gorkem · x · 2026-11-10
fal and Black Forest Labs launched FLUX 3 Max, a post-trained video model that scores 1,139 Elo in Artificial Analysis's preliminary image-to-video arena (tied #1, ahead of Gemini Omni Flash 1.1, Wan 3.0, H3 Max). It generates a 10-second 720/768p clip in 12.45 seconds — about 22x faster than Seedance 2.5 (277.8s) at equal quality. Served on the same API as H3 Max with one-model-ID switching. All generations are 75% off through November 22, making it the most affordable frontier video model at its quality level.
More from Multimodal
- Fable 5.2 generates a voxel-style lake scene worth a look — Angaisb_ · 2026-11-10
- Lightweight Muse Ember one-shots an interactive 3D tea garden, Meta reshares test — AIatMeta · 2026-11-10
- 2-step LoRA slashes LTX 2.5 video renders to 2 steps with SoL refiner — TimeTruth2490 · 2026-11-10
- Artificial Analysis hosts first image and video model evaluation event on Nov 17 in SF — ArtificialAnlys · 2026-11-10
- An end-to-end guide to training an image LoRA — bfl_ai · 2026-11-10
- Developer recreates Cartesia's blog experiment, visualizing model router chunk maps on images — davidwromero · 2026-11-10