Flux 3 is Black Forest Labs' newest image model on NightCafe. It launched in October 2026 as a single PRO model.
It is the next generation after Flux 2: native 4K output, prompt-to-edit, up to 10 reference images (included in the output price), consistent characters, and accurate text you can change in place.
Independent Artificial Analysis image arenas have not published Flux 3 scores yet (as of 6 October 2026). On NightCafe, treat Flux 3 as a premium high-resolution option — not the everyday 1-credit pick. For most prompts, start with GPT Image 2.5 Flare on Quality Low (~1 credit) or Muse Image (~2 credits).
In Studio, pick Flux 3 from the model picker. Write a prompt, optionally add start images, choose Resolution, then Generate.
You can generate up to 4 images at a time. Resolution defaults to Medium (about 1K).
Write a clear brief: subject, action, setting, camera, and any exact text in quotation marks. For an edit, add the image as a Start Image and describe only the change.
A square product photo for an online shop. A ceramic pour-over coffee dripper in matte speckled cream sits on a walnut board, with a linen cloth and a small bowl of coffee beans. Soft window light from the left. The kraft card in front reads "HOUSE BLEND". Photoreal, clean background.
Change the card text to "SINGLE ORIGIN" and keep the dripper, board, lighting, and camera position the same.
Credits are per image and depend on Resolution. Reference images are included in that price.
| Resolution | About | Credits |
|---|---|---|
| Low | ~0.75K (768×768 square) | 4 |
| Medium (default) | ~1K (1024×1024 square) | 4.75 |
| High | ~2K (2048×2048 square) | 10 |
| Ultra | ~4K (4096×4096 square) | 60.75 |
Flux 3 is not an unlimited model. Higher-cost settings typically need Fast Credits.
See also Flux 2 on NightCafe and the NightCafe AI Models Guide.