Black Forest Labs, the German startup behind some of the most popular open-source image generators, has released FLUX 3, its first model capable of producing video—and the same underlying system is reportedly being used to train robots on an Audi assembly line.
From Stills to Motion
The launch marks a significant pivot for the company, which built its reputation on high-quality still-image generation. With FLUX 3, Black Forest Labs is stepping into the increasingly competitive text-to-video arena, where rivals have raced to deliver realistic, prompt-driven clips.
The move reflects a broader industry trend in which generative AI developers are expanding beyond static images toward video, a far more demanding domain that requires models to maintain consistency across frames while simulating motion, lighting, and physics.
The leap from generating a single frame to generating coherent motion is one of the hardest problems in generative AI today.
For a European lab operating in a field dominated by well-funded American players, the release positions Black Forest Labs as a serious contender rather than a niche image specialist.
Teaching Robots to Move
Perhaps the more striking claim is that the same system powering FLUX 3 is being applied to robotics. According to the company, the technology is being used to help train robots working on an Audi assembly line—including guiding robotic hands through physical tasks.
The idea underscores a growing conviction among AI researchers that models trained to understand and predict visual sequences can double as "world models," giving machines a sense of how objects and environments behave over time. That capability is essential for robots operating in the physical world.
Bridging generative video and industrial robotics suggests Black Forest Labs sees its technology as more than an entertainment or content tool. Key takeaways from the announcement include:
- FLUX 3 is the lab's debut video-generation model.
- The same system is being deployed in robotics training.
- Applications reportedly extend to an Audi manufacturing environment.
If the robotics integration proves durable, it could signal a new phase for generative AI—one where the line between creating media and controlling machines continues to blur.