Google Lens Visual Match and Reverse Image Search avatar

Google Lens Visual Match and Reverse Image Search

Pricing

from $0.49 / 1,000 results

Go to Apify Store
Google Lens Visual Match and Reverse Image Search

Google Lens Visual Match and Reverse Image Search

Search Google Lens with image URLs or key-value store images and get visual matches, exact matches, product offers, detected entities, related searches and OCR text with page links, thumbnails, dimensions and prices.

Pricing

from $0.49 / 1,000 results

Rating

0.0

(0)

Developer

Xyrelith Data

Xyrelith Data

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

4 days ago

Last modified

Features

Limited permissions

Share

Search any picture with Google Lens and get the web pages, images and product offers behind it. Send public image URLs or images from the run's key-value store, pick the result set you need, and receive flat rows with page links, thumbnails, full-size image URLs, dimensions, prices, ratings and stock labels.

What you get

For every image the actor saves:

  • Visual matches — visually similar pages and images, in Google's order, with position, title, source site, domain, page URL, thumbnail, full-size image URL and pixel dimensions.
  • Product offers — price as shown plus parsed amount and currency, merchant, rating, review count and stock label for matches that are shoppable listings.
  • Exact matches — the pages carrying the same image, with title, source, publication date where Google shows one, the source image's pixel dimensions, a thumbnail and the resolved destination URL.
  • Google's AI description — the text Google's Lens result page writes about the image.
  • Detected entity — the subject name parsed from Google's own description, with the full description beside it.
  • Related searches — the follow-up questions Google suggests for the image, when the result page shows them.
  • Detected objects — the regions Lens found in the picture (whole image, main subject, objects, text blocks) with normalized boxes.
  • OCR text (optional) — the text read from the image, its language, and paragraphs with normalized boxes.
  • Image summary — one row per image with counts, status, warnings and the original image's dimensions.

Input

FieldTypeDescriptionDefault
modeselectall, visualMatches, exactMatches, products or ocr.all
imageUrlsarrayPublic direct image URLs. One or many.
imageKeysarrayOptional. Keys in this run's key-value store that hold image files.
countryselectGoogle market used for results and prices.US
languageselectResult language.en
maxItemsintegerMaximum match rows per image (1–1000).100
includeOcrbooleanAdd OCR text to the image summary for any mode.false
rehostImagesbooleanCopy full-size match images whose host blocks hotlinking into this run's key-value store and return a public URL, keeping the source URL in imageUrlOriginal.true

Example:

{
"mode": "all",
"imageUrls": [
"https://images.unsplash.com/photo-1502602898657-3e91760cbb34?w=640"
],
"country": "US",
"language": "en",
"maxItems": 50,
"includeOcr": false
}

Output

Each dataset row is a match or an image summary.

{
"recordType": "match",
"matchType": "visual",
"position": 1,
"inputImageUrl": "https://example.com/photo.jpg",
"title": "A guide to flying to France | Documents, Distance, Airports",
"source": "Sherburn Aero Club",
"domain": "sherburnaeroclub.com",
"pageUrl": "https://sherburnaeroclub.com/blog/a-guide-to-flying-to-france",
"thumbnailUrl": "https://encrypted-tbn0.gstatic.com/images?q=tbn:...",
"imageUrl": "https://sherburnaeroclub.com/storage/medialibrary/flying-to-france.jpg",
"imageWidth": 3530,
"imageHeight": 5309,
"searchCountry": "US",
"searchLanguage": "en",
"scrapedAt": "2026-11-15T00:00:00+00:00"
}

Product matches add price, currency, currencyCode, priceText, merchant, rating, reviewCount and inStock. Exact matches add date, googleRedirectUrl and the source image dimensions. When a full-size match image had to be copied into storage, imageUrlOriginal keeps the source URL. The image summary row adds aiOverview, knowledgeEntity, relatedSearches, detectedObjects, ocr, counts and status.

Modes

ModeWhat it returns
allVisual matches, product offers, AI description, entity, related searches and detected objects.
visualMatchesVisual matches and product offers only. Faster than all.
exactMatchesPages carrying the same image, up to Google's 400-row source ceiling.
productsOnly visual matches that carry a price.
ocrOnly OCR text, language and paragraph boxes. No browser session needed.

Add includeOcr: true to any non-OCR mode to attach OCR to the image summary.

Use cases

  • Find where a photo appears on the web and who published it.
  • Check whether your product photography is being reused by other sellers.
  • Turn a product photo into a list of similar listings with prices and stock.
  • Read text from screenshots, signs, labels and documents without a separate OCR tool.
  • Enrich a catalogue with visually similar pages and the entity Google recognizes.
  • Feed an AI agent structured visual-search results instead of raw HTML.

How it works

The actor prepares each image the same way Google's own upload widget does, sends it to Google Lens, and reads the result page's embedded match map plus its visible cards. Exact-match destination links are resolved from Google's redirect wrapper without visiting the merchant site. OCR uses Google's Chromium Lens endpoint. No Google account, cookie or API key is required.

Limitations

  • Visual matches are capped by Google. One lookup returns about 60 visual matches. A larger maxItems does not create more source results.
  • Exact matches are capped at 400 and can be empty. Google returns an exact set for some images and nothing for others; the same image can change between lookups.
  • The separate Products tab was retired by Google in October 2026. Product offers now arrive inside visual matches, so products filters price-bearing visual matches instead of opening a Products tab.
  • Product fields are shown only when the retailer publishes them. Price, rating, review count and stock are omitted when Google does not display them.
  • A separate Knowledge Graph entity card is no longer shown. The actor returns Google's AI description and derives the subject name from that description; when no name can be parsed, only the description is returned.
  • Related searches appear as AI follow-up questions when Google shows them. The current Lens page does not always include them.
  • Image URLs must be public and direct. Login-protected links, private hosts and web pages that merely contain an image are rejected.
  • Full-size match images belong to their source sites. Thumbnails are Google-hosted and always public. With rehostImages on (the default), images whose host blocks hotlinking are copied into the run's key-value store and returned as public URLs; imageUrlOriginal keeps the source link. An image that cannot be fetched at all is omitted rather than replaced by a broken link.
  • Google sometimes asks for a captcha. The actor retries with new sessions and connections; if Google refuses every attempt, the image summary says so and no match rows are saved for that image.
  • OCR is a best-effort read. Stylized text, very small type and low-contrast images may produce little or no text.

FAQ

Do I need a Google account or API key? No.

Can I search an image stored on my computer? Yes. Put the file in the run's key-value store and pass its key in imageKeys.

Why did an exact-match lookup return nothing? Google does not have an exact set for every image, and the set changes over time. Try the same image later or use visualMatches.

Why are some product fields missing? Google only shows a price, rating, review count or stock label when the listing publishes it.

Can I get more than 60 visual matches? No. That is Google's source ceiling, not an actor setting.

What does status mean on the image summary? ok means the result set was collected. partial means a secondary part (for example the AI description or exact tab) was unavailable. blocked means Google refused the Lens page for every connection. error means the image itself could not be loaded.

Are thumbnails and full-size image URLs usable outside Apify? Thumbnails are public Google-hosted images. Full-size URLs point at the source site and may be protected by that site.

Data source

Results come from public Google Lens pages. This actor is independent and is not affiliated with, endorsed by, or sponsored by Google LLC. Google Lens is a trademark of Google LLC.