Skip to main content
Nimble SERP is a realtime API that returns live Google search results as structured data. Submit a query and get back organic listings, pagination, and related questions, parsed and typed with high success rates.

​
SERP SLA

Default SLA and limitations for SERP API
Need higher throughput? SERP scales up to 1,000 QPS. Contact Sales to discuss your requirements.

​
Quick Start

​
Example Request

​
Example Response

​
How it works

1

Provide a query and search engine

Supply the search query, target engine (google_search), and optional geo-targeting
2

Nimble runs the search

Runs via Nimble’s infrastructure with geo-targeting applied, no rendering delays, no blocking
3

Results returned as structured data

Data is returned as parsed entity arrays under data.parsing.entities, with optional raw HTML when "html" is included in formats

​
Parameters

​
string
required
The search engine to query.
​
string
The search query string. Required for google_search, google_aio, google_news, google_images, and google_maps_search. Not used for google_maps_place or google_maps_reviews (use place_id instead).Example: "NBA Allstars 2026"
​
string
Google-assigned identifier for a specific place. Required for google_maps_place and google_maps_reviews. Cannot be used together with data_id in the same request.Example:
​
string
Alternative Google identifier for a specific place. Use instead of place_id when you have a data_id. Cannot be used together with place_id in the same request.Example:
​
object
Optional center point for google_maps_search. Biases results around the given location.Fields:
  • latitude (string)
  • longitude (string)
Example:
​
string
Run the search as if from a specific country. Returns geo-localized results - Optimized for USUse ISO Alpha-2 codes: US, GB, DE, FR, JP, etc.Example: "US"
​
string
Language preference for the search results. Use LCID standard.Examples:
  • en (English)
  • en-US (English, United States)
  • fr-FR (French, France)
​
string
Location context for the search. Accepted for google_search and google_images. Not supported for Maps engines.You can pass either:
  • A location string: Nimble resolves it to the corresponding UULE value (e.g. "United States", "New York, NY")
  • A raw UULE value: Google’s encoded location parameter (e.g. "w+CAIQICIiChIJOwQ...")
Examples:
  • "United States"
  • "London, England"
​
string
Filter results by time range. Only supported for google_search and google_images.Options:If omitted, no time filter is applied.
​
boolean
When set to true, removes the data.html field from the response. Useful for reducing response size when raw HTML is not needed.Default: false
​
string
Defines the order in which reviews are returned. Supported on google_maps_reviews only.
​
integer
default:"10"
Number of results to return. Supported on google_search, google_aio, and google_maps_reviews.
​
integer
Offsets the results by this number of listings. For example, start: 20 skips the first 20 results and returns the next page.Example: 20
​
integer
Offsets the results by this number of listings. Used for pagination with google_maps_search.Example: 20
​
boolean
default:"false"
When true, returns raw HTML per individual result page instead of a single stitched HTML string. Only relevant when num_results exceeds 10. Supported on google_search and google_images.Default: false
​
boolean
default:"false"
When true, increases the number of paid (sponsored) ads in the results by running the request in incognito mode. Requires JS rendering.Default: false
​
string
Emulate a specific device type. Supported on google_search and google_images.If omitted, defaults to desktop.
​
string
default:"com"
Google domain TLD to target. Supported on google_aio. Allows targeting country-specific Google domains.Examples: com, co.uk, de, fr, co.jpDefault: com
​
string
Cursor token for fetching the next page of reviews. Obtained from the previous google_maps_reviews response. Supported on google_maps_reviews only.

​
Search Engines

Returns organic web results, pagination, related questions, top stories, answer boxes, and more. Supported parameters:
  • query — search query string (required)
  • num_results — number of results to return, 1–100, default 10
  • start — result offset for fetching beyond the first batch
  • country — target country for localized results (ISO Alpha-2, e.g. US)
  • locale — language preference (LCID, e.g. en)
  • location — physical location context (string or raw UULE value)
  • time — filter results by recency (hour, day, week, month, year)
  • include_pages_html — return per-page HTML instead of a single stitched string
  • ads_optimization — boost sponsored results using incognito rendering (requires JS)
  • device — emulate a specific device type (mobile)
  • no_html — omit data.html from the response
cURL

​
Google AI Overview

Returns Google search results with the AI Overview block rendered at the top. The page is rendered to capture Google’s AI-generated answer alongside standard organic results. Supported parameters:
  • query — search query string (required)
  • num_results — number of results to return, 1–100, default 10
  • start — result offset for fetching beyond the first batch
  • country — target country, defaults to US (ISO Alpha-2)
  • locale — language preference (LCID, e.g. en)
  • location — physical location context (string or raw UULE value)
  • time — filter results by recency (hour, day, week, month, year)
  • domain — Google TLD to target (e.g. com, co.uk)
  • device — emulate a specific device type (mobile)
  • include_pages_html — return per-page HTML instead of a single stitched string
  • no_html — omit data.html from the response
Page rendering is always enabled and cannot be disabled. Expect higher latency than google_search.
cURL

​
Google News

Returns recent news articles matching the query, sourced from Google News. Supported parameters:
  • query — search query string (required)
  • country — target country for localized results (ISO Alpha-2, e.g. US)
  • locale — language preference (LCID, e.g. en)
  • no_html — omit data.html from the response
Pagination not supported. location and time parameters have no effect on this engine.
cURL

​
Google Images

Returns image results for a query, including image URLs, titles, and source pages. Supported parameters:
  • query — search query string (required)
  • start — result offset for fetching beyond the first batch
  • country — target country for localized results (ISO Alpha-2, e.g. US)
  • locale — language preference (LCID, e.g. en)
  • location — physical location context (string or raw UULE value)
  • time — filter results by recency (hour, day, week, month, year)
  • include_pages_html — return per-page HTML instead of a single stitched string
  • ads_optimization — boost sponsored results using incognito rendering (requires JS)
  • device — emulate a specific device type (mobile)
  • no_html — omit data.html from the response
ads_optimization requires JS rendering.
cURL
Returns business and location listings from Google Maps. Optionally bias results around specific coordinates. Supported parameters:
  • query — search query string (required)
  • coordinates — GPS coordinates to bias results around (latitude,longitude)
  • offset — result offset for paginating through listings
  • country — target country for localized results (ISO Alpha-2, e.g. US)
  • locale — language preference (LCID, e.g. en)
  • no_html — omit data.html from the response
Raw HTML is never returned for this engine — no_html is always true regardless of input.
cURL

​
Google Maps Place

Returns detailed information about a specific place by its Google Maps place_id. Supported parameters:
  • place_id — Google Maps place identifier (required if data_id not provided)
  • data_id — Google Maps data identifier (required if place_id not provided)
  • no_html — omit data.html from the response
Pagination not supported. Returns a single result per request.
cURL

​
Google Maps Reviews

Returns user reviews for a specific place by its Google Maps place_id. Supported parameters:
  • place_id — Google Maps place identifier (required if data_id not provided)
  • data_id — Google Maps data identifier (required if place_id not provided)
  • sort — review sort order (relevant, newest, highest_rating, lowest_rating)
  • num_results — number of reviews to return, range 10–20
  • paging_token_id — cursor token from previous response for next page
  • no_html — omit data.html from the response
num_results is capped at 20.
cURL

​
Response

All responses share the same top-level envelope:

​
Entities by engine

OrganicResult: one entry perweb resultAnswerBox: featured snippet at the top of resultsRelatedQuestionRelatedSearchPagination
NewsResults: the fullnews results object
ImageResult: one entry perimage
SearchResult: one entry perbusiness or location
Place: detailed information about a single place
Review: one entry peruser review

​
Async & Batch Requests

​
Parameters

​
string
Storage provider for results. When specified, results are saved to your cloud storage instead of Nimble’s servers.Options: s3 (Amazon S3), gs (Google Cloud Storage)
​
string
Bucket path where results will be saved. Results are stored as {task_id}.json at the specified location.Format: s3://your-bucket/path/prefix/Example: s3://my-bucket/nimble-results/
​
boolean
default:"false"
Compress results with GZIP before saving. Reduces storage costs and transfer time. When true, results are saved as {task_id}.json.gz.
​
string
Custom filename for the stored object instead of the default task ID.Example: "my-custom-name" saves as my-custom-name.json
​
string
Webhook URL to receive a POST request when the task completes. Nimble sends task metadata (without result data) to this URL when extraction finishes.Example: https://your-api.com/webhook/complete
Cloud storage delivery requires bucket permissions to be configured before use. See Bucket Permissions for setup instructions.

​
Async API

Submit a request and receive a task_id immediately — results are delivered via polling, webhook callback, or directly to your cloud storage bucket.
cURL
Response
Use the task_id to poll for status or retrieve results.
Polling, webhooks, and cloud storage delivery are all supported. See Callbacks & Delivery for the full guide.

​
Batch API

Submit multiple SERP requests in a single call. Use inputs for per-request parameters and shared_inputs for parameters shared across all tasks.

​
Examples

Search for different queries in one batch:
cURL
Per-request country and locale overrides. Shared inputs set the engine and delivery:
cURL
Set the query in shared_inputs and vary search_engine per input:
cURL

​
Batch response

Each task in the batch gets its own id. Results are saved to storage_url and a callback fires per completed task.

​
Check batch progress

cURL
Response
progress ranges from 0 to 1. When completed is true, all tasks are done.

​
List all batches

cURL
Response
Paginate by passing cursor until pagination.hasNext is false.

​
Retrieve batch summary

Once complete, fetch full task details for the entire batch:
cURL
Response

​
Pagination

google_news and google_maps_place do not support pagination.
1

Pull up to 100 results in one request

Set num_results: 100 to get up to 100 organic results in a single call. For most use cases, this eliminates pagination entirely.
cURL
2

Need more than 100? Add start to fetch the next batch

Combine num_results: 100 with start to pull 100 results at a time beyond the first page. Increment start by 100 for each subsequent batch.
cURL
start: 100 → results 101–200, start: 200 → results 201–300, and so on.
To receive raw HTML per result page instead of a single stitched HTML string, add "include_pages_html": true to your request.

​
Limitations

​
Next steps

API Reference

Full API reference for the SERP endpoint

Fast SERP

Lower latency variant optimized for time-sensitive workloads

Extract

Extract structured content from any result URL