Black Forest Labs releases FLUX 3 Action, a 7B open-weights world action model that tops RoboLab at 42.92%
Hugging Face Blog (Black Forest Labs)·medium signal
FLUX 3 Action is a 7B diffusion transformer. It takes a camera frame and a text instruction and denoises future video frames and the next two seconds of actions together, using Qwen3-VL-4B as the text encoder. Fine-tuned on DROID, it scores 42.92% on RoboLab-120, ahead of NVIDIA's 16B Cosmos3-Nano-Policy (36.8%) and pi0.5 (28.0%). The DROID and SO-101 checkpoints ship in LeRobot under the FLUX Kommunity License, so hobbyist arms can run it today.