Black Forest Labs

Black Forest Labs is a research and development organization focused on visual intelligence—specifically, building artificial intelligence models that can understand, reason about, and act upon visual information. The lab’s primary offering is the FLUX family of generative image models, accessible through an application programming interface (API) for developers, businesses, and researchers. These models are designed to convert text prompts into high-resolution images, with an emphasis on photorealistic output, fine-grained control, and production-level consistency. The FLUX.2 series includes three variants: FLUX.2 (the general-purpose model), FLUX.2 [klein] (a smaller, faster version optimized for lower latency use cases), and FLUX.2 [max] (a larger model with the highest fidelity and detail). Each model can be used to generate original images from scratch, edit existing visuals, or maintain character and style across multiple generations—a feature the lab terms "identity that holds." This consistency is critical for tasks such as creating branded content, generating product shots, or producing storyboards where the same subject must appear in different scenes without visual drift. A standout capability of FLUX.2 is its physics-aware rendering. When placing a product into an generated scene, the model automatically adjusts lighting, shadows, reflections, and perspective to match the environment, producing results that are difficult to distinguish from photographed images. This makes the technology suitable for e-commerce catalog creation, advertising mockups, architectural visualization, and concept art where realism and contextual accuracy are required. Beyond image generation, the lab provides pre-trained model weights for open‑weight research under a permissive lice