Esc
↑ ↓ to moveEnter to openEsc to close

AI tool, 10 free a day, 30 signed in

Image to prompt

To turn an image into a prompt, open the image here, keep Detailed prompt or pick another style, and press Write the prompt. Our AI describes the subject, setting, lighting, colors and style in words you can paste into an AI image generator. It can also write keyword tags, alt text or a plain description. The image is sent only when you press the button, and we keep neither it nor the words.

More text tools

How to turn an image into a prompt

  1. Open the image

    Choose a photo, drawing, render or screenshot, drop it on the box, or press Ctrl + V (Cmd + V on a Mac) to paste one.

  2. Pick what to write

    Detailed prompt is a full paragraph; Short prompt is one line. Keyword tags, Alt text and Description are there too.

  3. Press Write the prompt

    A smaller copy of the image goes to our AI, which describes what it sees and sends the words back.

  4. Copy and use it

    Copy the words, change any you like, and paste them into an image generator, a web page or a document.

What an image to prompt generator does

An image to prompt generator looks at a picture and writes the words an AI image generator would need to make something like it: the main subject, where it is, how the shot is framed, the light, the colors, the mood and the style, such as a photo, a watercolor or a 3D render. People use it to recreate a look they like, to start from a reference image instead of a blank box, to learn how good prompts are put together, and to keep a set of images in one style.

The same reading of the image gives four other useful results. Short prompt is a one-line version. Keyword tags is a comma-separated list for search, stock sites or tag-based generators. Alt text is one sentence for the alt attribute of an image on a web page. Description is a few plain sentences for someone who cannot see the picture.

Prompts, tags, alt text or a description: which to pick

Each style answers a different need:

The five styles and what each is for
StyleWhat you getUse it for
Detailed promptOne paragraph of about 60 to 120 wordsRecreating an image as closely as an image generator can
Short promptOne line of about 15 to 30 wordsA quick starting point you build on yourself
Keyword tags15 to 30 comma-separated words and phrasesTag-based generators, stock photo keywords, your own image library
Alt textOne sentence of up to 125 charactersThe alt attribute of an image on a web page or in an app
Description3 to 5 plain sentencesCaptions, notes, and readers who cannot see the image

What it can and cannot do

The words describe what can be seen. An image generator given them will make a new image in the same spirit, not a copy: the same prompt gives a different picture each time, and different generators read prompts differently. Expect to adjust a few words to get closer.

People are described only by what they wear, do and look like. The AI does not name anyone, does not guess who a person is, and does not guess age, ethnicity or other personal traits. Text in the image is quoted only when it can be read.

Everything is written in English. Small details in a busy picture can be missed, and the AI can be wrong about what something is, so read the result before you use it. Alt text in particular should say what matters on your page, which only you know.

Get a better prompt

Use a clear, uncropped image at a good size: before sending, your browser shrinks it to at most 1,600 pixels on its longest side, which keeps the main subject sharp. Crop away anything you do not want described with the crop tool, and blur faces you do not want sent with blur faces.

To find out whether an image you found was made with AI in the first place, check it with the AI image detector. To copy the words written in an image instead of describing it, use image to text.

Questions

How do I turn an image into a prompt?

Open the image on this page, keep Detailed prompt, and press Write the prompt. In a few seconds you get a paragraph that describes the subject, setting, lighting, colors and style, ready to copy into an AI image generator.

Is this image to prompt generator free?

Yes. Image to prompt shares the free AI image reads of the text tools: 10 a day without signing up, or 30 a day when you sign in with Google, which is free. Each image counts as one, and the count resets at midnight UTC.

Can it write alt text for an image?

Yes. Choose Alt text and you get one plain sentence of up to 125 characters that says what the image shows. Read it before you use it: good alt text also says why the image is on your page, which only you know.

Can it describe an image for me?

Yes. Choose Description for three to five plain sentences about what the image shows, where, who or what is in it and what they are doing, the colors and mood, and any text that can be read.

Will the prompt recreate my image exactly?

No. An image generator makes a new picture from the words, so you get something in the same spirit, not a copy, and each run is different. Adjust a few words to steer it closer.

Does it identify the people in a photo?

No. People are described only by what can be seen, such as their clothes and what they are doing. The AI does not name anyone or guess who they are, their age or their background.

Is my image uploaded or kept?

Only when you press Write the prompt. Your browser first makes a smaller copy of the image without its location or camera details, then sends it to the AI service we use, which describes it and sends the words back. We keep neither the image nor the words.

Why is there a daily limit?

Each read runs a paid AI model, so a daily limit keeps the tool free for everyone. Signing in with Google, which is free, raises it to 30 a day.

See all tools