Skip to main content
CreativoAI

Image to prompt

Upload an image and get a structured text-to-image prompt describing what is visible: subject, setting, composition, lighting, palette and style.

Image input2 credits per run

1. Your input

Image (JPEG, PNG or WebP, up to 5 MiB)

Drag an image here

or

Preview of your selected image

Your image is resized and re-encoded in your browser (this removes metadata), processed transiently and not stored. It is sent to our AI provider to create the result. Privacy details.

Best-effort wording presets, not exact syntax for any generator.

Checking your free credits…

Calculating the price…

You don’t have enough free credits left for this tool today. They reset at the time shown above.

2. Your result

Working on it…

This usually takes a few seconds. You can cancel, but the AI service may still finish work it already started.

AI can make mistakes. Check results before you use them.

Sample output for a bundled example photo (not generated live)

main prompt
A ceramic bowl of oranges on a rustic wooden table beside a sunlit window, soft diffused light, shallow depth of field, warm natural palette, editorial food photography.
subject
Bowl of oranges
lighting
Soft window light from the left (interpretation)
Saved on this device (optional)

Off by default. When on, text results (never images) are kept in this browser only, up to 20 items for 7 days. They are not sent to us.

About this tool

Turn a reference image into words. The result describes what can be seen and offers wording presets for popular generators. It is a description, not a recovered original prompt: the tool cannot know the seed, model, artist or camera used.

Limitations to keep in mind

  • It cannot recover the real prompt, seed, model or camera metadata.
  • Style and lens wording are interpretations of what is visible.
  • Generator presets are best-effort wording, not guaranteed syntax.
  • Results can contain mistakes; read and edit before using.

Frequently asked questions

Does this give me the original prompt?

No. It writes a new prompt that describes what is visible. Any camera or style wording is an interpretation.

Which image formats work?

JPEG, PNG and WebP up to 5 MiB. Large images are resized in your browser before upload.

Is my image stored?

It is processed transiently and sent to the AI provider to produce the result. We do not keep it. See the privacy page for details.

Why a negative prompt only sometimes?

Only some workflows, such as Stable Diffusion-style prompts, use negative prompts, so it is shown only for that target.