Google Images Scraper is a free and open-source scraper that gets you unlimited detailed Google Images data for free.
- 🖼️ 300 images per search — full-size links, width & height, file size, source page & site
- 🎛️ Every Google Images filter — size, colour, type, aspect ratio, file type, usage rights, time & site
- 🏷️ Licences, credits & product prices — stock-photo licences, photographer credits, shop prices & sellers
- 🔁 Related & reverse image search — 300 look-alike images per image, plus every page carrying a picture
{
"search_information": { "query": "golden retriever puppy", "total_results": 114000000 },
"images": [
{
"position": 1,
"id": "eYoKU9jW-eRT6M",
"title": "Golden Retriever: Pros, Cons & Diet Guide | Ollie",
"link": "https://blog.ollie.com/wp-content/uploads/250506_Livingroom_SHOT-10_0173_FR-scaled.jpg",
"width": 2560,
"height": 1707,
"file_size": "522KB",
"format": "jpg",
"source": { "name": "Ollie Blog", "domain": "blog.ollie.com", "page_link": "https://blog.ollie.com/breed-deets-golden-retriever/" }
},
{
"position": 22,
"id": "32eplErtBrE7rM",
"title": "Cuddly Golden Retriever Puppies Are the Cutest Thing You'll See Today - A-Z Animals",
"link": "https://a-z-animals.com/media/2025/10/shutterstock-2300276251-medium-licensed.jpg",
"width": 1000,
"height": 667,
"is_licensable": true,
"credits": { "credit_line": "Shutterstock", "creators": ["ConstanzaMartinez"] },
"license": { "provider": "shutterstock.com", "details_link": "https://www.shutterstock.com/image-photo/2300276251?utm_source=iptc&utm_medium=googleimages&utm_campaign=image" }
}
],
"suggested_filters": [{ "text": "Newborn", "query": "newborn golden retriever puppy" }],
"related_searches": [{ "query": "chocolate golden retriever puppy" }]
}Trimmed for readability.
1️⃣ Clone and install (the last command downloads Camoufox, the hardened Firefox that opens Google for you):
git clone https://github.com/omkarcloud/google-images-scraper
cd google-images-scraper
python -m pip install -r requirements.txt
python -m camoufox fetch2️⃣ Start the API:
python run.py3️⃣ Get your first data:
curl "http://localhost:8000/images/search?query=golden+retriever+puppy"{
"count": 114000000,
"per_page": 100,
"current_page": 1,
"next": "http://localhost:8000/images/search?query=golden+retriever+puppy&page=2",
"search_information": { "query": "golden retriever puppy", "total_results": 114000000 },
"images": [
{
"position": 1,
"id": "jUS0xT_ulT-SUM",
"title": "How to Train a Golden Retriever Puppy: Growth & Training Timeline",
"link": "https://s3.amazonaws.com/cdn-origin-etr.akc.org/wp-content/uploads/2020/07/09151754/Golden-Retriever-puppy-standing-outdoors.jpg",
"width": 729,
"height": 486,
"file_size": "190KB",
"format": "jpg",
"source": { "name": "American Kennel Club", "domain": "www.akc.org", "page_link": "https://www.akc.org/expert-advice/dog-breeds/golden-retriever-puppy-training-timeline/" }
}
]
}All 5 endpoints are now live at http://localhost:8000.
The first search can take a minute while a browser opens Google and clears its checks; after that, each call takes a few seconds. Autocomplete needs no browser and answers in a blink.
Google often answers a fresh browser with an "unusual traffic" captcha. The scraper first tries on its own — no key, no cost — and when Google insists, it tells you so. Then:
- Get a CapSolver API key (one captcha is solved per ~30 calls).
- Start the API with it:
CAPSOLVER_API_KEY=CAP-... python run.py
Google pauses one IP after a few dozen searches, so for real volume add a sticky proxy too ({session} gets a fresh id for every browser session):
GOOGLE_IMAGES_PROXY="http://user-session-{session}:pass@host:port" CAPSOLVER_API_KEY=CAP-... python run.py5 endpoints cover everything you need.
| Endpoint | Path | Returns |
|---|---|---|
| Search Images | /images/search |
100 images per page with full-size links, sources, licences & prices |
| Autocomplete | /images/autocomplete |
Search suggestions, exactly as Google's search box shows them |
| Image Details | /images/details |
Everything about one image, plus 12 related images |
| Related Images | /images/related |
100 visually similar images per page |
| Reverse Image Search | /images/reverse-search |
Every page carrying a picture, or look-alike images |
The same API is published on RapidAPI, and its playground is the easiest place to try parameters and see raw responses. Once a request looks right, run it locally for unlimited free data.
- Subscribe to the free plan — 200 calls/month, no credit card.
- Try the endpoints in the playground — every param is pre-filled, so you see real data in one click.
- Copy the generated code and replace
https://best-google-images-scraper-free-200-calls.p.rapidapi.comwithhttp://localhost:8000. It will now run against your local API.
import requests
# generated by the playground, host swapped for the local API
response = requests.get(
"http://localhost:8000/images/search",
params={"query": "golden retriever puppy", "size": "large", "usage_rights": "creative_commons"},
)
print(response.json())You're a developer — we know how hard completing a project can be. So we offer full support: just message us and we'll reply ✅ with a solution within 1 working day.
- Google Maps Scraper (3,100+ GitHub Stars) — type "dentists in New York", get every business as a ready-to-call lead list: phones, emails, websites & reviews. Up to 100K free leads/month.
- G2 Scraper — G2 product details, ratings & AI-found contacts
- Website Email Contact Scraper — emails, phones & socials from any website
- AliExpress Scraper — live product details, SKU variants, stock & shipping
- Booking Scraper — Booking.com hotels: prices, ratings, rooms & amenities
- Etsy Scraper — Etsy products: prices, discounts, shops & variations
⭐ Love It? Star It ⭐!
Star the repo ⭐ and become my star hero!
It's just 1 click, but it means the world to me.
