# Changelog of Bing Reverse Image Search 🔍 Similar Images, Pages & OCR (`dz_omar/bing-reverse-image-search`) Actor

- **URL**: https://apify.com/dz_omar/bing-reverse-image-search/changelog.md
- **Full Actor documentation**: https://apify.com/dz_omar/bing-reverse-image-search.md

## Changelog

### 0.0 — 2026-11-08

First release.

- Reverse image search on Bing from an image link, an uploaded file or a `data:` link. Several pictures in one run.
- Bing's best guess for every picture, plus the search it suggests and the search that best represents it.
- Visually similar images, most similar first: full-size URL, thumbnail, Bing's cached copy, width and height, file size, file type, publish date, date discovered, how many pages carry the image, and the page it sits on. A single picture can return several hundred.
- The pages that carry the picture itself, with the same fields.
- Related and shopping searches, text written in the picture (with where it sits) and Bing's AI description with numbered sources.
- Any JPG, PNG, WebP, GIF, BMP or TIFF up to about 25 MB: large pictures are resized for you, transparent ones are flattened, phone photos are rotated the right way up.
- 18 markets and three SafeSearch levels.
- A limit per picture and result type, or everything Bing returns.
- A picture is charged once and only when Bing returns something for it; a match is never delivered or charged twice. A maximum cost per run is respected.
- Pictures that cannot be used (a web page instead of a picture, a missing file, too small, too large) leave a single notice row and never fail the run; nothing is charged for them.
- If Bing refuses a connection, the Actor asks again on a fresh one before giving a picture up.
- Interrupted runs resume where they stopped, without repeating or re-charging rows.
- Long runs stop themselves shortly before the run's time limit and leave a notice, instead of timing out.