Amidst the thriving AI landscape in San Francisco, a lean 70-person startup headquartered 5,000 miles away in Germany’s Black Forest, Black Forest Labs, has emerged as a top competitor to Silicon Valley’s leading AI labs in image generation. In December, Black Forest Labs secured funding at a $3.25 billion valuation, following deals to power AI image-generation features for Adobe and the graphic design platform Canva.
The company has also forged agreements with major AI labs such as Microsoft, Meta, and xAI to integrate similar capabilities into their products. Nearly two years post-launch, Black Forest Labs can now be selective about its partnerships. Notably, in 2024, Elon Musk’s xAI initially tapped Black Forest Labs to power Grok’s first image generator. This partnership, while raising the startup's profile, also generated controversy due to the chatbot’s limited safeguards and concluded months later when xAI developed its in-house AI image model.
More recently, xAI reportedly approached Black Forest Labs again to license its technology, but sources familiar with the matter indicate that Black Forest Labs declined. The company deemed it too operationally challenging to partner with xAI, known for its famously chaotic work environment. Furthermore, in September, Black Forest Labs inked a significant $140 million multiyear deal to grant Meta access to its AI image-generation technology.
The appeal of Black Forest Labs for these major AI entities stems from its image generators being among the world's best, ranking just below offerings from OpenAI and Google on Artificial Analysis' third-party benchmarks. The startup also provides some of the most downloaded text-to-image models on Hugging Face, suggesting that numerous free AI image tools on the market are likely powered by a version of Black Forest Labs’ technology.
This performance is particularly impressive given the company’s historically fewer resources compared to its competitors. This constraint led them to pioneer a more efficient research path known as latent diffusion. This technique involves an AI model first sketching a rough blueprint of an image, then progressively adding detail. Co-founder Andreas Blattmann stated that latent diffusion “enabled us to put out very powerful models that took orders of magnitude less resources than our competitor’s models.”
Despite its current success, Black Forest Labs views image generation as merely the beginning. Blattmann revealed plans for the startup to unveil a robot powered by one of its AI models later this year, though the hardware manufacturer remains undisclosed. This initiative is part of a broader vision to build AI capable of perceiving and acting within the physical world, emphasizing the expansive potential of visual intelligence.