Using Describe a Picture
This service accepts one JPG, PNG or WebP picture at a time. The homepage produces English descriptions and alt text. The AI Image Describer also answers single image questions, extracts visible text in its original language and creates English visual prompts. Text extraction does not translate and does not guarantee complete transcription of dense or long documents. It is free to use without an account. Submit only images, questions and context you are permitted to send for third-party processing.
Shared availability
The service allows up to 7,000 generation requests per UTC day across all visitors and both tool pages. Browser and IP daily limits, rolling minute limits, a browser concurrency limit and bot verification also apply. New generations, regenerated versions and manually retried requests count separately. Model requests that fail or time out also count. The service may be unavailable during provider or database failures.
Do not attempt to bypass usage controls, impersonate other visitors or interfere with the service. There is no payment or registration option to unlock more requests.
Review the output
Descriptions, answers, extracted text and prompts may contain mistakes or omit information. Compare them with the original picture before use. Alt text should be reviewed in the context of the page where the image appears. This tool does not guarantee accessibility compliance, search rankings or accurate transcription of small text.
Page-session results
Pictures and results remain in the current page session. The service does not provide a saved history or recovery of lost results. See the Privacy Policy for processing details.