POST /images
Google Images API
The BestSERP Google Images API runs a live Google Images search and returns each image as JSON: the full-size image URL, its width and height, Google's thumbnail, the title, and the page and site that host it. Ask for up to 5 scroll loads per request, at one credit each.
Last updated:
Key takeaways
- /images returns images with title, imageUrl, imageWidth, imageHeight, thumbnailUrl and more.
- Target any country with gl and any language with hl; city-level location is not supported yet.
- 1 credit per page of results ($0.50 per 1,000 after 3,000 free); failed searches are refunded.
- POST a JSON body or GET with the same parameters; the key goes in the X-API-KEY header.
Try the Google Images API
Pick a suggested query to see a saved response, or type your own for a live search (3 free per day).
Loading the live demo
Send a request
POST a JSON body to
https://api.bestserp.com/images with your key in the X-API-KEY header. Replace YOUR_API_KEY and run.curl -X POST 'https://api.bestserp.com/images' \
-H 'X-API-KEY: YOUR_API_KEY' \
-H 'Content-Type: application/json' \
-d '{"q":"golden retriever"}'Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
qrequired | string | Search query (1–256 characters). | |
gl | string | us | Country, 2-letter code (us, gb, vn, de…). The search runs from an IP in that country when one is free, else from another one (see ipCountry). |
hl | string | en | Interface language (en, vi, de…). |
pages | integer | 1 | Result pages or scroll loads to fetch, 1–5. Each costs 1 credit. |
tbs | string | Recency: qdr:h (hour), qdr:d (day), qdr:w (week), qdr:m (month), qdr:y (year). Also accepted: timeRange = hour|day|week|month|year. | |
noCache | boolean | false | Skip the 1-hour result cache (5 minutes for news). |
async | boolean | false | Answer 202 immediately with a requestId to poll. |
timeout | number | 60 | Seconds to wait before answering 202 (1–120). |
autocorrect | boolean | Accepted for compatibility; no effect. |
Response fields
Results are in
images. Empty fields are left out.images[]
| Field | Meaning |
|---|---|
position | Rank of the image, from 1. |
title | Title of the image or its page. |
imageUrl | URL of the full-size image. |
imageWidth | Width of the full-size image, in pixels. |
imageHeight | Height of the full-size image, in pixels. |
thumbnailUrl | Google's thumbnail of the image. |
source | Site name of the page that hosts the image. |
domain | Host name of that page, without www. |
link | URL of the page that hosts the image. |
Every response
| Field | Meaning |
|---|---|
searchParameters | Your request after validation, with defaults filled in (q, gl, hl, type, page or pages...). |
requestId | ID of this request. Poll GET /requests/{requestId} for async requests and for final links. |
credits | Credits charged for this request (refunded automatically if it fails). |
cached | true when the answer came from the recent-result cache (1 hour, 5 minutes for news). |
linksPending | true while Google redirect links are being replaced by final URLs; poll the request to get them. |
unresolvedLinks | Links that are still Google redirects: they work, but are not the final URL. |
durationMs | Time the search took, in milliseconds. |
ipCountry | Country of the IP the search ran from (same code as gl). Another country than gl when no IP there was free: the search runs anyway instead of waiting. Missing when unknown. |
Sample response
A trimmed response for "golden retriever". Your results reflect Google at the time of the request.
{
"searchParameters": {
"q": "golden retriever",
"gl": "us",
"hl": "en",
"type": "images",
"pages": 1,
"engine": "google"
},
"images": [
{
"position": 1,
"title": "Golden retriever puppy in the grass",
"imageUrl": "https://pupphotos.com/images/golden-puppy.jpg",
"imageWidth": 1600,
"imageHeight": 1067,
"thumbnailUrl": "data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg' width='320' height='200'><rect width='320' height='200' fill='%23f5c26b'/></svg>",
"source": "Pup Photos",
"domain": "pupphotos.com",
"link": "https://pupphotos.com/golden-puppy"
},
{
"position": 2,
"title": "Golden Retriever: breed characteristics and care",
"imageUrl": "https://dogbreedguide.com/images/golden-breed.jpg",
"imageWidth": 1200,
"imageHeight": 800,
"thumbnailUrl": "data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg' width='320' height='200'><rect width='320' height='200' fill='%23e9b35a'/></svg>",
"source": "Dog Breed Guide",
"domain": "dogbreedguide.com",
"link": "https://dogbreedguide.com/golden-breed"
},
{
"position": 3,
"title": "Golden retriever running on the beach",
"imageUrl": "https://happytails.blog/images/golden-beach.jpg",
"imageWidth": 1920,
"imageHeight": 1280,
"thumbnailUrl": "data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg' width='320' height='200'><rect width='320' height='200' fill='%23f7d9a0'/></svg>",
"source": "Happy Tails",
"domain": "happytails.blog",
"link": "https://happytails.blog/golden-beach"
},
{
"position": 4,
"title": "Two golden retrievers playing",
"imageUrl": "https://petdaily.com/images/golden-two.jpg",
"imageWidth": 1080,
"imageHeight": 1080,
"thumbnailUrl": "data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg' width='320' height='200'><rect width='320' height='200' fill='%23f0c47a'/></svg>",
"source": "Pet Daily",
"domain": "petdaily.com",
"link": "https://petdaily.com/golden-two"
}
],
"requestId": "5b0c8a52-3f0e-4d55-9a2a-0f3a8e2d1c11",
"credits": 1,
"cached": false,
"linksPending": false,
"unresolvedLinks": 0,
"durationMs": 2410
}What you can build
Visual research
Collect reference images for a topic or product, with the page each one comes from.
Brand and image monitoring
See which sites show your product photos or logo for a query.
SEO tools
Check how a site's images rank in Google Images for target keywords.
SEO tools guide
Current limits
What this API does not do yet, so you can decide before you build on it.
- Image search filters (size, color, type, usage rights) are not supported yet; only the recency filter (tbs) is.
- Up to 5 scroll loads (pages) per request; each costs 1 credit.
- Targeting is by country (gl) and interface language (hl). City-level location is not supported yet: a request with location is refused with an explicit error.
- Results are what a desktop browser sees. There is no mobile SERP option yet.
- 60 searches per minute per account by default; ask on Telegram for more.
Pricing: 1 credit per page of results, 3,000 free credits, then $0.50 per 1,000. See pricing
Frequently asked questions
Do I get the full-size image or only a thumbnail?
Both: imageUrl is the full-size image and thumbnailUrl is Google's thumbnail, with imageWidth and imageHeight of the full image.
How many images does a request return?
The images Google loads for one scroll of the results. Set pages up to 5 to load more; each load costs 1 credit.
Can I filter by image size or color?
Not yet. You can limit results by recency with tbs (qdr:d, qdr:w, qdr:m...).
Can I use the images I find?
The API tells you where images are published; it does not grant any rights to them. Check the license on the source page before reusing an image.
Other Google APIs
- Google Search APIOrganic results with titles, links and snippets, plus the video carousel and Top stories.
- Google News APINews articles with source, publish time and site icon, from the Google News tab.
- Google Videos APIVideo results with duration, channel and platform.
- Google Shopping APIProducts with price, numeric price, discount, delivery and rating.
- Google Places APILocal businesses with rating, category, address and opening hours.
- Google Maps APIGoogle Maps places with coordinates, place id and CID.
Start with 3,000 free searches
No credit card. Sign in with Google, create a key, send your first request.