FLUX 3 Image AI image generator

FLUX 3 Image is the image generation and editing model in the FLUX 3 family from Black Forest Labs, the lab behind the FLUX models. It generates images from text, edits an image by written instruction, combines up to ten reference images, and accepts bounding boxes that fix where each element goes. Flux 3 Image, this site, is an independent workbench that gives you that model in the browser and bills per image in credits. We are not affiliated with, endorsed by, or sponsored by Black Forest Labs.

Sign in with Google, type what you want, pick an aspect ratio and a resolution, and hit generate. The credit price sits on the prompt box before you run, and the result lands in the same panel and in your history. Everything happens in the browser: no API key, no install, nothing to configure.

FLUX 3 Image by Black Forest Labs, accessed through fal. Text prompts use the text-to-image endpoint; attaching a reference selects the edit endpoint. There is one model, not a basic and a premium tier; the resolution determines the credit price shown on the prompt box.

Three resolutions: 1K (about one megapixel), 2K (about four), and 4K (about sixteen). Fourteen aspect ratios: 21:9, 2:1, 16:9, 3:2, 7:5, 4:3, 5:4, 1:1, 4:5, 3:4, 5:7, 2:3, 9:16, and 1:2, plus Auto, which follows your reference image or falls back to square.

Attach one reference (up to 10 MB, at least 256 pixels on each side and at most 4 megapixels) and write the change as an instruction: change the jacket to blue, remove the cat, swap the plate for a burger. Name the element precisely and say what should stay unchanged. The workbench sends your instruction and reference to fal's FLUX 3 Image Edit endpoint.

Yes. Accurate text rendering is one of the things Black Forest Labs built FLUX 3 Image for, including multilingual text. Put the exact wording in quotation marks and say where it goes. Large headlines are the most reliable; very small type and long paragraphs are still worth checking before you ship.

Yes. Add bounding boxes to the end of your prompt: each box is [top, left, bottom, right] on a 0 to 1000 grid with a short description of its element. The model places each element inside its box. On an existing image the same boxes target an edit, so you can recolor, move, replace, or remove one element while everything else stays unchanged. Our guide on bounding boxes walks through example prompts.

It depends on the resolution: 44 credits at 1K, 91 at 2K, and 550 at 4K. Reference images and edits cost the same as a fresh generation at the same resolution. The exact price sits on the prompt box and follows the size you pick, so there is never a surprise. A failed or timed-out generation refunds its credits automatically; you only pay for images you actually get.

Every new account starts with 5 free credits, no card required. They are not enough for an image on their own (a 1K image is 44 credits), but they never expire and count toward your first image once you add a plan or a pack. Paid plans open once pricing is finalized; your signup credits stay on the account until then.

Monthly plan credits stay valid 30 days from each grant. Yearly plans hand you all twelve months upfront, valid for a year, so unused credits carry over month to month. One-time packs carry their own validity window, shown on the pack. Subscriptions are self-serve: cancel anytime under Settings → Billing and keep the credits already granted.

Every paid plan includes commercial use, and we add no visible watermark or logo to your downloads. You stay responsible for what you put in the frame: trademarks, real people's likenesses, and brand assets follow their own rules regardless of which tool made the image.

Yes. Every generation is tied to your account and appears only in your own history; nothing is posted to a public feed or gallery. Your prompt and any reference you upload go to Black Forest Labs' API solely to render that image, and you can ask us to delete your data at any time.