Describe Image
Generates a concise description / alt-text for an image using a hosted vision model. Output is plain text. Uses 1 credit per run.
About Describe Image
Describe Image generates a short, plain-English description of any photo using a hosted vision model, ideal for alt text and accessibility. Reach for it when you have a batch of images that need captions for a website, a CMS, or a screen-reader-friendly document. It returns plain text and costs 1 credit per run.
- Category
- export
- Input
- Accepts: image/png, image/jpeg or image/webp.
- Output
- Outputs: text/plain.
- Cost
- Credit-metered
- Memory
- low
Common uses
- Write alt text for product photos so your store meets accessibility guidelines
- Caption blog header images without stopping to phrase each one yourself
- Generate descriptions for an image library so it becomes keyword-searchable
- Add screen-reader text to social posts before scheduling them
- Draft a starting caption you can lightly edit instead of writing from scratch
- Describe user-uploaded photos in a moderation or cataloging workflow
Frequently asked questions
What formats can I describe?
PNG, JPEG, and WebP. The description comes back as plain text ready to paste into an alt attribute or caption field.
How long is the description?
It's intentionally concise, a sentence or short phrase suited to alt text rather than a paragraph. For a richer, multi-sentence description, use the detailed in-browser captioner.
Does the image get uploaded?
Yes. This Pro tool sends the image to a hosted vision model. If you'd rather nothing leaves your device, the free in-browser captioners do this offline after a one-time model download.
Is the caption always accurate?
It's usually solid for everyday photos, but vision models can miss small details or misread unusual scenes. Skim the output before publishing it as alt text.
What does it cost?
1 credit per image.
Keywords
- describe
- caption
- alt-text
- accessibility
- pro
- vision