FLUX.1 is a new family of AI image generation models released by Black Forest Labs — founded by the original creators of Stable Diffusion. It represents the state-of-the-art in open-source image generation, widely recognized as matching or exceeding Midjourney v6 in image quality, photorealism, and prompt adherence. The FLUX.1 family offers three variants: FLUX.1 [pro] (highest quality, API only), FLUX.1 [dev] (open weights for non-commercial use), and FLUX.1 [schnell] (fastest, fully open Apache 2.0 license). This tiered approach serves both enterprise API users and the open-source community. FLUX.1's text rendering is among the best in AI image generation — it can accurately render sentences, signs, labels, and complex typographic compositions within images, a capability where most models fail. Its human anatomy generation is also notably superior, with accurate hands, fingers, and proportions that have historically been weakness areas. The model architecture uses a novel hybrid of multimodal and parallel diffusion transformer blocks, allowing significantly higher image quality with better computational efficiency. Resolution and aspect ratio support are flexible, with native high-resolution generation. FLUX.1 has been integrated into major platforms including Replicate, ComfyUI, fal.ai, and is available via the Black Forest Labs API. Its combination of open weights (for dev/schnell) and exceptional quality has made it the new reference point for open-source image generation.