How to use it
- 1
Add an image. Drop it in or paste a screenshot.
- 2
Read the description. The AI writes a one-line caption, which you can edit.
- 3
Copy it. Copy just the alt text, or the full
<img>tag for your HTML.
[ADD YOUR OWN TEST HERE: what you tried, screenshots of the result, what worked and what didn't, and the date.]
Why alt text matters
Screen readers read alt text aloud to people who can't see the image, and search engines use it to understand what an image shows. Images without alt text are invisible to both. Most accessibility guidelines require it on every meaningful image.
How to write good alt text
- Describe what's important, not every detail: "Red running shoe, side view" beats "a picture of a shoe".
- Don't start with "image of" or "picture of".
- Add what only you know: the product name, the person, the place.
- Decorative images (dividers, backgrounds) should have empty alt text:
alt="".
How it works and its limits
The tool runs ViT-GPT2, an open-source image captioning model, in your browser. It gives a general, short description. It can't read text in images (use Image to text for that), recognize specific people or brands, or judge context, so treat it as a first draft.
Questions
Is it free with no sign-up?
Yes.
Can it describe images in other languages?
It writes in English. Use our AI translator for other languages.
Does it upload my images?
No. The model runs on your device.
How long should alt text be?
Usually one short sentence, under about 125 characters.