AI Image to Prompt Generator

The AI Image to Prompt Generator turns any picture into prompts you can paste into an AI image generator to make one like it. Upload a photo, a piece of AI art or a screenshot, and get a prompt for ChatGPT and Gemini (Nano Banana), one for Midjourney, one for Stable Diffusion and Flux with a negative prompt, a video prompt for Veo, Kling and Seedance, and a short one-liner.

Why one prompt doesn't fit every app

ChatGPT and Gemini understand plain sentences; Midjourney works best with short descriptive phrases and parameters like --ar; Stable Diffusion and Flux read comma-separated tags and take a separate negative prompt; video models need motion and a camera move. The same picture needs a differently written prompt for each, so you get all of them from one upload.

What you get

• Prompts for every major app, ready to copy.
• A breakdown of the subject, medium, camera, light, colour and mood, so you can change one part and keep the rest.
• Style keywords that define the look.
• Remixes: more cinematic, anime, photo, watercolour, 3D, or your own change.

How to use it

1. Upload a picture, or choose An idea and describe it.
2. Choose Recreate this picture, or Use its style for my subject and write your subject.
3. Click Get prompts and copy the one for your app.

Tips for better AI images

• Lead with the subject, then the setting, style, light and camera.
• Name the medium: "35mm film photo", "watercolour", "3D render" changes more than any adjective.
• Light is the look. "Golden hour", "neon rim light" or "overcast soft light" sets the mood.
• Change one thing at a time when you refine a result.

Related tools

Sharpen a picture with the Image Enhancer, crop it to the shape you need with the Image Cropper, or write a caption for it with the AI Instagram Caption Generator.

Frequently asked questions

How do I turn an image into a prompt?

Upload the picture and click Get prompts. The AI looks at the subject, style, camera, light and colour and writes prompts that recreate it, one for each kind of AI image app, so you can paste the right one straight into ChatGPT, Gemini, Midjourney or Stable Diffusion.

Which AI image generators do the prompts work in?

The natural prompt is for ChatGPT (GPT Image), Gemini (Nano Banana), Copilot, Ideogram and Leonardo. There is a Midjourney prompt with --ar set to your picture's shape, a Stable Diffusion and Flux prompt with a negative prompt, a video prompt for Veo, Kling and Seedance, and a short prompt for apps like Canva.

Will the new image look exactly like mine?

Close, not identical. A prompt describes a picture; the model paints a new one each time. For a closer match, attach your original picture in ChatGPT or Gemini as well and ask for "this style". Faces of real people are always described, never recreated.

Can I use the style of a picture for my own subject?

Yes. Choose "Use its style for my subject" and write your subject, for example "my dog on a skateboard". The prompts keep the picture's medium, light, colour and composition and put your subject in it.

I don't have a picture. Can it write a prompt from an idea?

Yes. Choose "An idea", write a sentence and pick a look and shape. You get the same set of prompts, with the setting, light and camera filled in.

Is it free?

Yes. Every day you get free AI runs, and one run gives the whole set of prompts. Copying and the examples on this page are always free. A remix uses one run.