Google SERP API

Google search results as structured JSON, from Python or the command line: organic results with the AI Overview and related searches, Google News, Google Images, local places, autocomplete suggestions and Google Trends. The code is free and open source (MIT). It runs on the QuanticData Data API, so there is no Google API key to request and no proxy pool to maintain: sign in once and run a query.

pip install git+https://github.com/quantumproxies/google-serp-api
google-serp login
google-serp search "best usb c hub for macbook" --country us --num 20 -o serp.json

login opens the QuanticData sign-in page in your browser. New accounts get $2 of free usage every month, no card required: create one.

What it scrapes

Command	Input	Returns	Per run
google-serp search	query	Rank, title, link, snippet, date, sitelinks, 9 fields per row (docs and price)	up to 100
google-serp news	topic	Title, publisher, date, snippet, link, 6 fields per row (docs and price)	up to 100
google-serp images	query	Title, thumbnail, source page, host, 5 fields per row (docs and price)	up to 100
google-serp places	business type + --location	Name, rating, reviews, category, address, phone, website, coordinates, hours, 15 fields per row (docs and price)	20 per page, more with --page
google-serp autocomplete	seed query	Suggestion, relevance score, type, 4 fields per row (docs and price)	the whole suggestion list
google-serp trends	keyword	Date, interest value (0 to 100), 4 fields per row (docs and price)	one year of weekly values

Any country and language: --country us --lang en, --country de --lang de. --location "Milan, Italy" returns the page a searcher in that city sees.

Sample output

A real row from google-serp search "best usb c hub for macbook" --country us --lang en --num 10 (2026-10-22):

{
  "rank": 1,
  "title": "Best USB-C hubs and dongles: Add ports to your laptop or tablet",
  "link": "https://www.pcworld.com/article/402858/the-best-usb-c-hubs-for-your-laptop-tablet-or-2-in-1.html",
  "display_link": "www.pcworld.com › Accessories › Computer Accessories",
  "source": "pcworld.com",
  "description": "Anker 555 8-in-1 USB-C Hub Best overall USB-C hub. Excellent build quality Supplies 85W of power Good mix of ports. USB-C is common among Macs, ...",
  "date": "Oct 7, 2026"
}

That is one entry of organic. With -o results.json you get the whole response, which also carries search_metadata, search_parameters, related_searches, ai_overview, pagination.

Python

from quanticdata_google_serp_api import GoogleSERP

serp = GoogleSERP()  # saved login, or QUANTICDATA_API_KEY

result = serp.search("best usb c hub for macbook", country="us", num=20)
for r in result:
    print(r["rank"], r["source"], r["title"])
print(result.meta.get("ai_overview"))
print([q["query"] for q in result.meta.get("related_searches", [])])

news = serp.news("electric vehicles", country="us")
shops = serp.places("coffee shop", location="Austin, Texas, United States")
ideas = serp.autocomplete("google serp")
interest = serp.trends("air fryer", country="us")
interest.to_csv("air_fryer_trend.csv")

Each call returns a Run: iterate it for the rows (plain dicts), take len(), or save it with to_csv(), to_json(), to_jsonl() or to_dataframe() (pandas). Errors are specific: InsufficientCredit, RateLimited, InvalidInput, JobFailed, AuthError.

Examples

• Rank tracking: Record where a domain ranks for a list of keywords, per country, and append the positions to a CSV every day. See examples/rank_tracker.py.
• AI Overview monitoring: See which of your keywords trigger an AI Overview and keep its text, to follow how Google answers your topics. See examples/ai_overview_watch.py.
• Trend research: Weekly Google Trends interest for a list of keywords in one long-format CSV, ready for a chart. See examples/trends_to_csv.py.

Command line

Results print as a table; -o file.csv, -o file.json or -o file.jsonl writes a file. List inputs take a b c, a,b,c, @file.txt (one per line) or - (stdin). With -o results.json you get the whole API response; the table, CSV and JSONL views hold its main list.

google-serp search

Google web results for a query: organic results with rank, title, link, snippet and date, plus the AI Overview, related searches and pagination when Google shows them.

google-serp search QUERY [options]

Option	Type	Default	Description
QUERY	string		What to search, e.g. "best usb c hub for macbook".
--country	string		Country to run the request from, as an ISO 3166-1 alpha-2 code (us, gb, de...).
--lang	string		UI language, e.g. en, it.
--location	string		Search from this place, e.g. "Milan, Italy": the results a searcher there sees.
--device	desktop / mobile	desktop	Desktop or mobile results page.
--num	integer		How many organic results to collect, 1 to 100; Google serves about 10 per page and the pages are merged.
--page	integer		Result page, 1-based.
--safe	active / off		SafeSearch.

Sent with every call: engine=google, search_type=search.

google-serp news

Google News results for a query: title, publisher, how long ago, snippet and link.

google-serp news QUERY [options]

Option	Type	Default	Description
QUERY	string		Topic to search in Google News, e.g. "electric vehicles".
--country	string		Country to run the request from, as an ISO 3166-1 alpha-2 code (us, gb, de...).
--lang	string		UI language, e.g. en, it.
--location	string		Search from this place, e.g. "Milan, Italy": the results a searcher there sees.
--num	integer		Results to request (clamped to 1–100 by the service; Google serves ~10 per page and merges pages — see search_metadata.paging).
--page	integer		Result page, 1-based.

Sent with every call: engine=google, search_type=news.

google-serp images

Google Images results for a query: thumbnail, the page that uses the image and its host.

google-serp images QUERY [options]

Option	Type	Default	Description
QUERY	string		What to search in Google Images.
--country	string		Country to run the request from, as an ISO 3166-1 alpha-2 code (us, gb, de...).
--lang	string		UI language, e.g. en, it.
--num	integer		Results to request (clamped to 1–100 by the service; Google serves ~10 per page and merges pages — see search_metadata.paging).
--page	integer		Result page, 1-based.
--safe	active / off		SafeSearch.

Sent with every call: engine=google, search_type=images.

google-serp places

Local results for a query in a place: name, rating, review count, category, address, phone, website, coordinates, opening hours and the place id.

google-serp places QUERY [options]

Option	Type	Default	Description
QUERY	string		Business type or name, e.g. "coffee shop"; set the place with --location.
--location	string		Where to search, as a place name, e.g. "Austin, Texas, United States".
--country	string		Country to run the request from, as an ISO 3166-1 alpha-2 code (us, gb, de...).
--lang	string		UI language, e.g. en, it.
--num	integer		Results to request (clamped to 1–100 by the service; Google serves ~10 per page and merges pages — see search_metadata.paging).
--page	integer		Result page, 1-based.
--gps-coordinates	string		maps/places: map center lat,lon[,zoom]. Wins over uule and location.

Sent with every call: engine=google, search_type=places.

google-serp autocomplete

The suggestions Google offers while typing a seed query, with Google's own relevance score: real searches people make, for keyword research.

google-serp autocomplete QUERY [options]

Option	Type	Default	Description
QUERY	string		Seed query whose suggestions you want, e.g. "google serp".
--country	string		Country to run the request from, as an ISO 3166-1 alpha-2 code (us, gb, de...).
--lang	string		UI language, e.g. en, it.

Sent with every call: engine=google, search_type=autocomplete.

google-serp trends

Google Trends for a keyword: interest over time for the last 12 months (0 to 100), plus the top and rising related queries in the full response.

google-serp trends QUERY [options]

Option	Type	Default	Description
QUERY	string		Keyword to read in Google Trends, e.g. "air fryer".
--country	string		Country to run the request from, as an ISO 3166-1 alpha-2 code (us, gb, de...).
--lang	string		UI language, e.g. en, it.

Sent with every call: engine=google, search_type=trends.

Output fields

google-serp search: 9 fields

Field	Type	Description
rank	integer	Position on the results page, 1-based.
title	string	Result title.
link	string	Target URL.
display_link	string	Breadcrumb shown under the title.
source	string	Domain of the result.
description	string	Snippet shown under the title.
date	string or null	Date shown in the snippet, when present.
rating	number or null	Star rating shown in the snippet, when present.
reviews	integer or null	Review count shown with the rating.

google-serp news: 6 fields

Field	Type	Description
rank	integer	Position in the news results.
title	string	Headline.
link	string	Article URL.
source	string	Publisher.
date	string	Age or date as shown ("20 hours ago").
description	string	Snippet, when Google shows one.

google-serp images: 5 fields

Field	Type	Description
rank	integer	Position in the image grid.
title	string	Caption as shown in the grid.
thumbnail	string	Thumbnail URL.
source	string	Host of the page that uses the image.
source_url	string	Page that uses the image.

google-serp places: 15 fields

Field	Type	Description
rank	integer	Position in the local results.
name	string	Business name.
rating	number	Star rating.
reviews	integer	Review count.
category	string	Main category.
address	string	Address as shown.
website	string	Website host, when listed.
link	string	Google Maps link.
gps_coordinates	object	{ latitude, longitude }.
place_id	string	Place id, the input of place details and reviews.
data_id	string	Same id in the form the reviews call takes.
weekly_hours	object or null	Opening hours per weekday.
open_state	string	Open or closed now, as shown.
phone	string or null	Phone number, when listed.
description	string or null	Short description, when Google has one.

google-serp autocomplete: 4 fields

Field	Type	Description
rank	integer	Order in the suggestion list.
value	string	The suggested query.
relevance	integer	Google's relevance score for the suggestion.
type	string	Suggestion kind (QUERY for a plain query).

google-serp trends: 4 fields

Field	Type	Description
date	string	Week as shown by Google Trends.
timestamp	integer	Start of the week, Unix seconds.
value	integer	Interest, 0 to 100 relative to the peak of the period.
partial	boolean or null	True for the current, incomplete week.

Sign-in and API keys

• google-serp login signs you in with OAuth 2.1 and PKCE in the browser. The token is stored in ~/.config/quanticdata/credentials.json (readable only by you) and refreshed automatically. One login serves every QuanticData tool on the machine.
• No browser on the machine (SSH, Docker, Colab): run google-serp login --no-browser, open the link anywhere, then paste back the address the browser ends on.
• Scripts and CI: set QUANTICDATA_API_KEY to a key from the dashboard, or store one with google-serp login --api-key.
• google-serp logout revokes the token. The connection is listed under API keys in the dashboard as "API · QuanticData CLI"; deleting that key disconnects every machine that uses it.

Pricing

The code is free and MIT-licensed. Searches are billed to your QuanticData account per call, at the price shown on the SERP API page linked above; $2 of usage is free every month and failed calls are never billed. Plans with higher limits: pricing.

Other ways to call it

The same data over plain HTTP:

curl -X POST https://api.quanticdata.io/v1/serp \
  -H "Authorization: Bearer $QUANTICDATA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"engine": "google", "search_type": "search", "query": "best usb c hub for macbook", "country": "us", "lang": "en", "num": 10}'

• API reference with Python, Node, PHP and curl examples.
• MCP server to give Claude, Cursor or any MCP client the same tools.
• n8n node for no-code workflows.

FAQ

Is this Google SERP API free?

The code is free and MIT-licensed. Searches run on your QuanticData account: $2 of usage is free every month with no card, then you pay per search at the price on the SERP API page. Failed calls are never billed.

Do I need a Google API key or proxies?

No. There is no Google Cloud project, no Custom Search key and no proxy to configure: you send a query and get the parsed results page back.

Does it return the AI Overview?

Yes, when Google shows one for the query: it is in the ai_overview field of the full response (-o results.json in the CLI, result.meta in Python).

How many results can one search return?

Up to 100 organic results with --num; Google serves about 10 per page and the pages are fetched and merged for you. Use --page to address one specific page.

Can I search from a specific country or city?

Yes: --country and --lang set the Google market and language, and --location "Milan, Italy" returns the results a searcher in that city sees.

Can I export to CSV?

Add -o results.csv for the main list (organic results, news, images, places, suggestions or the Trends series); -o results.json keeps the whole response.

Related scrapers

• Google Maps Scraper: Google Maps scraper for Python and the command line.
• YouTube Scraper: YouTube scraper for Python and the command line.

License

MIT. Built and maintained by QuanticData.