Built for phones first
Most inspiration arrives on a phone: a screenshot from social media, a photo from a walk, an artwork in a gallery. The tool was designed for small screens first, with large tap targets, a one-tap copy button and results that appear as they are written, so you’re never left staring at a spinner.
Tap the upload area to pick an image from your photos, choose the target model, and tap the generate button. When the prompt is ready, copy it and switch to Midjourney, Gemini, ChatGPT or your favourite Stable Diffusion app.
Add it to your home screen
If you use it often, add AnyToPrompt to your home screen for one-tap access. On iPhone, open the site in Safari, tap the Share button and choose Add to Home Screen. On Android, open it in Chrome, tap the menu and choose Add to Home screen. It then opens like an app, without an app store download.
There’s no account to create, so you can start straight away on any device, including a tablet or a laptop, where you can also paste images with Ctrl or ⌘ + V.
Private by design
Many photo apps ask for access to your whole library. Here you choose exactly one image each time. It’s processed in memory to write your prompt and then discarded, and your prompt history stays on your own device. Nothing is stored on our servers. You can clear your history on the device at any time.
Every major generator, one tool
The app speaks six prompt formats: General, Midjourney, Stable Diffusion, Flux, Nano Banana and DALL-E. Switch formats and generate again to compare, and use any of 14 output languages. It’s free for up to 20 prompts a day. Pick a format once and it stays selected while you work through a batch of images.
Tips for better results on mobile
Crop your screenshot to the part you care about before uploading, since interface elements and captions can end up in the description. Photos taken in good light give richer colour and lighting details. On a slow connection, a smaller image uploads faster and produces the same quality of prompt, because large images are scaled down anyway. And when a result is close but not quite right, edit it right in the box before copying; you don’t need to generate again.
Guides from our blog
- Why your AI images look generic, and seven prompt habits that fix it
AI image prompts that say how a picture should look, not just what is in it: seven habits, with before-and-after examples, for sharper results.
- How to reverse-engineer any image into a prompt, and when it works
A practical reverse prompting workflow: describe an image, format it for your generator, and refine one element at a time. What works, and what never will.