AI modelsNovember 16, 2026
-22
š§
Black Forest Labs Bets Beyond Images
Black Forest Labs explains why FLUX 3 trains on video, image and audio together.
#Black Forest Labs#FLUX 3#video generation#world models#diffusion

š„ What happened
Black Forest Labs (BFL) launched FLUX 3, a model trained jointly on images, video, and audio. The team behind Stable Diffusion and SDXL is pivoting from pure image generation to multimodal world models.
š” Why it matters
BFL argues a video reveals how an object looks from behind ā a still only shows where it is. Its video upscaler shares the FLUX 3 backbone to kill frame-to-frame shimmer. That directly threatens specialized video tools.
ā” Our take
BFL owned the image market with FLUX ā now it's coming for motion and physics. Anyone optimizing only stills is already behind.
The title, summary and analysis of this item were produced automatically by an AI system and have not been editorially reviewed. They may contain errors, bias or omissions ā when in doubt, read the linked original source.
