Black Forest Labs brings FLUX to robotics
The image-model maker's FLUX 3 Action Base extends its generative stack into world-action modeling, released through the LeRobot ecosystem.
Black Forest Labs, best known for its FLUX family of image generators, has stepped into robotics with FLUX 3 Action Base, a world-action model published on Hugging Face. The model is distributed through LeRobot, the open robotics stack that has become a common home for policies, datasets, and control models.
The release signals an ambition beyond static image synthesis. A "world-action" model is designed to connect perception with control — taking in multimodal inputs and producing the kind of outputs that can drive a robot's behavior. That places FLUX 3 Action Base in the same broad category as the vision-language-action models that labs and startups have been racing to build.
Why it matters
- It extends the FLUX architecture, previously aimed at generative imagery, into embodied AI.
- Shipping via LeRobot lowers the barrier for researchers already working in that ecosystem to experiment with the model.
- It marks Black Forest Labs' entry into a fast-moving field dominated by robotics-first teams.
Details remain thin at launch: the record lists no published parameter count, context length, or benchmark figures, and the license is marked simply as "other," so would-be users should check the terms on the model card before building on it. As an initial release, it establishes a base that future iterations can build from.
For a company that helped define the current wave of open image models, the move into world-action modeling is a notable bet that the same generative foundations can help machines act, not just render.
Sources
- Visit
black-forest-labs/flux-3-action-base
Hugging Face
More from Black Forest Labs
All Black Forest Labs releases →Black Forest Labs ships FLUX.3 Image Klein for ComfyUI
The lab's latest open image model arrives packaged for ComfyUI with both generation and editing built in.
Black Forest Labs ships FLUX.3 Image Klein
The latest open image model from the FLUX team handles both text-to-image generation and editing, and arrives ready to run in Diffusers.
Ideogram 4.0 arrives as an open-weight image model
A 9.3-billion-parameter text-to-image model lands on GitHub with downloadable weights and code.
More in Any-to-Any
All Any-to-Any →
InternLM's Sage S2.1 Flash targets scientific reasoning
The 320B mixture-of-experts model is built to reason over scientific evidence and act inside interactive environments.
Haru-4B Brings Full-Duplex Speech to Six Languages
A compact 4-billion-parameter model handles real-time, two-way voice conversation across six languages in a single system.

Meituan Releases LongCat 2.5, an MIT-Licensed MoE
The Chinese delivery giant expands its open-model push with a multimodal, agentic mixture-of-experts system.
0 comments
No comments yet. Be the first to weigh in.