Black Forest Labs Unveils FLUX 3 Video Generation Model

Black Forest Labs has launched FLUX 3, its first multimodal foundation model. This video generation model supports text-to-video and image-to-video creation, generating clips up to 20 seconds in HD quality with native audio.

The company announced the launch on July 23, 2026, after working with Crusoe AI to provide the necessary hardware support. Early access to FLUX 3 became available through the BFL API and to a select group of partners in late July.

Black Forest Labs has been building towards robotics use cases, and FLUX 3's architecture is designed to feed into those pipelines. The company plans to release an open-weight version of the FLUX 3 Dev backbone at a future date.