Black Forest Labs: FLUX, Open Research, and the Future of Visual AI
AI Engineer Europe 2026 · 22:32
Generative AI models for images, video and audio
Black Forest Labs develops FLUX, a family of generative AI models for creators, developers and enterprises. Users can generate images from text, edit existing images and combine visual references into new designs. The company offers a browser playground, APIs for application integration and licensed model weights for deployment and customization on users’ infrastructure. FLUX 3 extends the family into video with native audio, offering clips up to 20 seconds in early access.
Founded in Freiburg in 2024, the company’s ten-person founding team includes CEO Robin Rombach and Andreas Blattmann. Their earlier research helped develop latent diffusion, the foundation for Stable Diffusion. BFL’s own FLUX.1 combined diffusion transformer blocks with flow matching to train image generators. Its subsequent Self-Flow approach underpins FLUX 3, aligning generation and understanding within one architecture trained jointly on images, video and audio, with extensions into action prediction for robotics.
In December 2025, the company reported tens of millions of Hugging Face downloads and partnerships with more than a dozen Fortune 500 enterprises; Adobe, Canva, Deutsche Telekom and Meta had integrated FLUX into their workflows. That year’s $300 million Series B, led by AMP and Salesforce Ventures, valued BFL at $3.25 billion post-money and brought total funding above $450 million.
AI Engineer Europe 2026 · 22:32
Affiliations reflect their AIE appearances, not necessarily current employment.
Batifol framed robotics and action prediction as directions for broader visual intelligence, presenting a SelfFlow robot-manipulation demonstration.
Affiliations reflect each recorded session, not necessarily current employment.