FLUX 1.1 Pro holds the highest overall Elo score on the Artificial Analysis image benchmark — outperforming Midjourney 6.1, Ideogram v2, and all other models in both visual fidelity and prompt accuracy.

Generates photorealistic images in 4.5 seconds — six times faster than the original FLUX.1 Pro. State-of-the-art inference speed without sacrificing quality.

12B parameter architecture with rotary positional embeddings and parallel attention layers delivers images that are indistinguishable from camera-shot photos — perfect for advertising and commercial production.

Ultra Mode generates images at up to 2048×2048 pixels — four times the resolution of standard models — with precise prompt following maintained at maximum resolution.

Raw Mode produces hyper-realistic images in the style of candid photography — perfect for lifestyle, editorial, and documentary-style visuals that demand authenticity.

Exceptional accuracy at interpreting complex prompts with specific requirements. Increased variety and creativity in outputs while maintaining consistency — ideal for precise creative briefs.