Black Forest Labs has launched FLUX 3, a multimodal flow model family that includes video generation capabilities with native audio. The release also introduces FLUX-mimic, a variant designed for video-action robotics applications.

  • FLUX 3 Video supports text-to-video, image-to-video, and video-to-video generation, along with keyframe-to-video transitions.
  • All outputs from the model include native audio generation capabilities.
  • The system features agentic chaining to create longer multi-shot sequences and supports multilingual dialogue.
  • Black Forest Labs also announced FLUX-mimic, which demonstrates the model's ability to drive robots and predict their impact in factory settings.

The company is preparing to release an open weights Dev version of FLUX 3 Video.