FLUX 3 Live on Together AI: Multi-Scene Generation with Synced Audio

togethercompute · x · 2026-08-07

Together AI announced the launch of FLUX 3, the new multimodal model from Black Forest Labs. The model can generate multiple scenes and camera angles in one go, featuring multilingual dialogue with lipsync, ambient sound, and control via text, images, or keyframes. It supports up to 20-second clips with synchronized video and audio.

Related event: Together AI Launches FLUX 3 Serverless API(2 posts)→

Original post →

More from Multimodal

Multimodal channel →