Skip to main content
GET
Google Light web search
A successful uncached request costs 10 credits. Failed requests and cache hits are free. Check X-Credits-Used for the actual charge and /api/public/pricing for current prices and promotions.
The JSON response contains search_information, organic_results, related_searches and pagination. Light skips ads, AI Overviews, Knowledge Graph, People Also Ask, local results, news, inline videos and shopping. No mode parameter is required. Passing mode=full or ai_overview=true cannot enable rich SERP features. Use start=10, 20, and so on for later pages. Google may return fewer results than requested by num; a page commonly contains about ten organic results. Set output=html to receive the raw Google response instead of parsed JSON. Light has an eight-second processing budget, including fetching, fallback and parsing. If the budget expires, the response is 503 with Retry-After. The CLI command is scrapebadger google light-search "vinted api". The MCP tool is google_light_search, with start for pagination. See the Light overview or regular Search.

Authorizations

x-api-key
string
header
required

Query Parameters

q
string
required

Search query (supports Google operators like site:, inurl:, intitle:)

gl
string | null

Country code (e.g. us, gb, de, fr). Defaults to the country implied by domain (e.g. google.co.uk → gb), falling back to us when domain doesn't carry a country hint.

hl
string
default:en

Language code (e.g. en, es, fr)

num
integer
default:10

Results per page

Required range: 1 <= x <= 100
start
integer
default:0

Pagination offset (0, 10, 20...)

Required range: x >= 0
domain
string
default:google.com

Google domain (e.g. google.com, google.co.uk)

device
string
default:desktop

Device target — selects the Google SERP layout via a matching User-Agent. One of: desktop, mobile, iphone, android, tablet.

user_agent
string | null

Custom User-Agent to send to Google. Overrides device. Use to target a specific browser/OS fingerprint.

Maximum string length: 512
Pattern: ^[^\r\n]*$
location
string | null

City-level geo-targeting (e.g. 'New York, USA')

lr
string | null

Language restrict (e.g. lang_en)

tbs
string | null

Time-based search filter (e.g. qdr:d for past 24h)

safe
string
default:off

Safe search (active/off)

uule
string | null

Encoded location parameter (UULE)

filter
integer | null

Include omitted results (0=show all, 1=filter)

Required range: 0 <= x <= 1
nfpr
integer
default:0

Disable auto-correction (1=exact match)

Required range: 0 <= x <= 1
cr
string | null

Country restrict (e.g. countryUS|countryGB)

ludocid
string | null

Google My Business CID (place ID)

lsig
string | null

Knowledge Graph map view ID

kgmid
string | null

Knowledge Graph entity ID

si
string | null

Cached search parameters

ibp
string | null

Layout control (e.g. gwp;0,7)

uds
string | null

Google filter string

output
enum<string>
default:json

Response format. json (default): structured parsed SERP. html: the raw Google SERP HTML exactly as fetched, with Content-Type: text/html and no parsing. Use html when you run your own parser. Pagination/geo/device/UULE params all still apply; the ai_overview chase is skipped for html output.

Available options:
json,
html

Response

Successful Response

Essential SERP data returned by GET /api/v1/light/search.

search_information
SearchInformation · object
required

Metadata about the search query.

pagination
Pagination · object
required

Pagination metadata.

organic_results
OrganicResult · object[]