Skip to main content

​
What is Prompting?

Prompting in text-to-image generation refers to the process of entering a text description — the prompt — that FLUX uses to generate a matching image. Your prompt is the primary way you communicate intent: what should be in the image, how it should look, and what mood or style it should convey.
Chromatic 3D cursor on a black background

a chromatic 3D rendition of a cursor, black background

​
What Can a Prompt Look Like?

A prompt passed to FLUX can take many forms. There is no single correct format — what matters is that your description gives FLUX enough to work with.
This template is a useful starting structure, not a strict formula. You can keep it concise, expand individual slots, or ignore parts that are not relevant to your image.

​
Use Natural Language

FLUX works best when your prompt reads like a clear description of the image you want to generate. Natural language helps the model understand what should appear in the image, how the elements relate to each other, and what visual direction to follow. The clearer the description, the easier it is for FLUX to produce a focused and consistent result.

​
Text in images

When you want FLUX to generate specific text inside an image, place the exact wording in quotation marks. This makes it clearer that the text should appear visibly in the final image, rather than being treated as part of the general prompt description. Quotation marks help separate written content from the rest of the scene, which gives FLUX a stronger signal to render the words as text.

​
Refine As You Go

Strong prompts usually come from iteration, not from trying to write the perfect prompt on the first attempt. A practical loop is:
  1. Start with a simple version
  2. Check what FLUX got right and wrong
  3. Adjust one important detail at a time
If the image is close but not there yet, tweak the subject, framing, lighting, or style before rewriting everything.

​
Multilingual Prompting

FLUX can be prompted in multiple languages. You don’t need to write in English to get great results — FLUX understands a wide range of languages and responds to them with the same level of quality.
That said, English prompts tend to produce the most precise results, as the majority of FLUX’s training data is in English.

​
Image Input

With FLUX.1 Kontext and FLUX.2, your prompt isn’t limited to text. These models accept up to 10 images as additional input alongside your text prompt — allowing you to edit existing images, transfer styles, maintain character consistency across generations, or composite multiple references into a single output.
Red and black butterfly in flight
Silver metallic butterfly in flight

The butterfly is now made of shiny silver

Image-based prompting is covered in depth in the Image Editing with FLUX section of this guide.