Black Forest Labs
3 storiesBlack Forest Labs develops the FLUX family of open generative AI models. Recent coverage has focused on its new releases: FLUX 3 Action, a seven-billion-parameter robotics model that predicts a robot's next action from camera feeds, and Flux 3 Image, which supports multi-step edits, bounding-box scene composition, up to ten reference images, and 4K output. It then released FLUX.3 Image Klein, supported day one in Diffusers.
Related topics
Black Forest Labs released FLUX.3 Image Klein, an open image generation model, with day-one support in Diffusers. It unifies text-to-image, multi-reference image-to-image, and bounding-box layout control. The distilled checkpoint generates a 1024x1024 image in about 2 seconds on an H100. It offers NVFP4 and FP8 weights, LoRA training scripts, and ships as a Modular Diffusers pipeline.
Black Forest Labs released Flux 3 Image, which supports multi-step edits without altering the rest of the image, scene composition with bounding boxes and up to ten reference images, and up to 4K output. Open weights are coming in the next few weeks.
Black Forest Labs entered robotics with FLUX 3 Action, an open-world-action model that uses camera feeds to predict a robot's next action. At seven billion parameters, it sets a record on the RoboLab-120 benchmark and runs up to 3.95x faster than the previous top model.
That is everything
The A.I. Daily
Black Forest Labs releases FLUX.3 Image Klein, available day-one in Diffusers