FLUX.1 is a state-of-the-art text-to-image model family created by Black Forest Labs.
Released on August 1 2024, the 12-billion-parameter rectified-flow-transformer set quickly established itself as a leading alternative to Midjourney, DALL·E 3 and Stable Diffusion-XL by offering faster inference, superior prompt-faithfulness and typography support.
Key features
- Edition choice – Schnell (Apache 2.0 open-weight), Dev (research, non-commercial) and Pro (commercial API) balance openness, speed and quality.
- In-context editing – the 2025 “Kontext” upgrade lets users supply both text and images, preserving characters and structure while applying incremental edits.
- Creative tool-chain – FLUX.1 Tools (Fill, Depth, Canny, Redux) add in-painting/out-painting and structure-guided control for precise workflows.
- Ecosystem reach – integrations with Replicate, Together AI, Nvidia Blackwell PCs and UIs such as ComfyUI, Krea and Freepik boost adoption.
By uniting flow-matching training with transformer scalability, FLUX.1 delivers photorealistic output, flexible aspect ratios up to 4 MP (Ultra mode) and end-to-end latencies as low as 10 s, setting a new benchmark for accessible, high-quality generative imaging.
