Black Forest Labs

Started as a PhD researcher at LMU Munich and Heidelberg University, where he co-invented Latent Diffusion, a generative modeling technique that led to Stable Diffusion. Before founding Black Forest Labs, he was a generative AI researcher at NVIDIA and Stability AI, contributing to subsequent versions of Stable Diffusion including Stable Diffusion 2.0 and SDXL. Co-founded Black Forest Labs in 2024 alongside Robin Rombach, Patrick Esser, and Dominik Lorenz. Publications have amassed over 22,000 citations, and was named one of Germany's Top 40 Under 40 in 2024. **Leadership trait:** His commitment to open-source principles has fostered a collaborative environment, accelerating innovation and accessibility in AI technologies.

Co-founder of Black Forest Labs, a German-American AI company founded in 2024 alongside Robin Rombach, Andreas Blattmann, and Patrick Esser—all key researchers behind the latent diffusion technology that powered Stable Diffusion. Lorenz contributed to the foundational latent diffusion research at Ludwig Maximilian University of Munich and Heidelberg University under Björn Ommer, then joined Stability AI to develop subsequent versions of Stable Diffusion, including Stable Diffusion 2.0 and SDXL. After Stability AI experienced financial difficulties and leadership turmoil in 2023-2024, the technical team departed to start Black Forest Labs, where he continues as research scientist working on scaling flow matching transformers for image synthesis.

A researcher from LMU Munich and Heidelberg University who developed latent diffusion and Stable Diffusion before joining Stability AI. As one of four original latent diffusion paper authors, Esser joined Stability AI to commercialize the foundational research. Left Stability AI during its 2023-2024 financial difficulties to co-found Black Forest Labs. Esser's research philosophy centers on the field's collaborative adoption—he measures success by seeing others build on his work—a principle evident in Black Forest Labs' open-weights approach to FLUX models.

Studied physics at the University of Heidelberg (2013–2020) before pursuing a PhD at LMU Munich. One of the original co-creators behind latent diffusion, adversarial distillation, and Stable Diffusion; later served as research director at Stability AI before departing in early 2024. Research has centered on making generative models more efficient, contributing to VQGAN, Latent Diffusion, Stable Diffusion and Flux. Assembled a core team of former Stability AI researchers for Black Forest Labs' 2024 launch. Prefers a low profile, with "his satisfaction coming from being acclaimed by the AI community" rather than public visibility.