Is this Google's Knowledge Graph Search API?

No. That API queries Google's entity database and returns IDs and schema.org types. SerpKite reads the knowledge panel on the live results page, so you get the description, facts and website people see in a search.

What if Google shows no panel?

knowledge_graph is absent and the call is billed like any search that returned results. Queries that name one entity ("eiffel tower", "marie curie", a company name) get panels most often; questions often get an answer_box instead.

Does it work in other languages?

Yes. Set country and language; the panel comes back in that language when Google has one for it.

POST /v1/search · fields=knowledge_graph

Google Knowledge Graph API

The knowledge panel Google shows next to an entity search, parsed into one object: title, type, description with its source, website and the panel's facts. Read from live results, so it matches what people see today.

Get a free API key Try in playground

1 credit per search 2,500 free credits, no card Failed calls are free

Overview

What the Google Knowledge Graph API does

When a search names a person, place, organisation or work, Google shows a knowledge panel. /v1/search returns it as knowledge_graph: title, type ("Tower in Paris, France"), description with description_source and description_link, website when the panel links one, and attributes, the panel's facts as label/value pairs ("Height": "984′, 1,083′ to tip", "Born": "November 7, 1867, Warsaw, Poland").

Add fields: "knowledge_graph" to get only the panel (plus request and meta), which keeps agent context small. Google's own Knowledge Graph Search API returns entity IDs and schema.org types from its database, not the panel on the results page; this endpoint reads the panel people actually see, with its facts.

What you get

knowledge_graph object: title, type, description, description_source, description_link, website and image_url when shown, and attributes (label → value). Absent when Google shows no panel.
request object: Echo of the normalized request (endpoint, engine, q, country, language, page…), handy for logging.
meta object: request_id, credits_used, cached, engine, latency_ms, resolved_urls and parse_quality. parse_quality=empty means the call was not billed.

Request

POST /v1/search

cURL

Python

TypeScript

fetch

PHP

curl https://api.serpkite.com/v1/search \
  -H "Authorization: Bearer $SERPKITE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"q":"marie curie","fields":"knowledge_graph"}'

Response (live response, trimmed)

200 OK · application/json

{
  "knowledge_graph": {
    "title": "Marie Curie",
    "type": "Polish-French physicist and chemist",
    "description": "Maria Salomea Skłodowska Curie, better known as Marie Curie, was a Polish and naturalised-French physicist and chemist. She shared the 1903 Nobel Prize in Physics with her husband, Pierre Curie…",
    "description_source": "Wikipedia",
    "description_link": "https://en.wikipedia.org/wiki/Marie_Curie",
    "attributes": {
      "Born": "November 7, 1867, Warsaw, Poland",
      "Died": "July 4, 1934 (age 66 years), Passy, France",
      "Awards": "Nobel Prize in Physics, Nobel Prize in Chemistry",
      "Spouse": "Pierre Curie (m. 1895–1906)",
      "Children": "Ève Curie, Irène Joliot-Curie"
    }
  },
  "request": {
    "endpoint": "search",
    "engine": "google",
    "q": "marie curie",
    "country": "us",
    "language": "en"
  },
  "meta": {
    "request_id": "req_01J8ZK4M6Q2V7",
    "credits_used": 1,
    "engine": "google",
    "cached": false,
    "latency_ms": 812,
    "resolved_urls": true,
    "parse_quality": "ok"
  }
}

Every response also carries X-Credits-Used, X-Credits-Remaining, X-Cache and X-Tokens-Estimate headers. Run this query in the playground

Reference

Parameters

The parameters /v1/search accepts, in the JSON body or the query string.

Google Knowledge Graph API parameters
Name	Type	Default	Description
`q` required	string	–	The search query. Required. Up to 2,048 characters.
`country`	string	us	Country to search from, as a two-letter ISO code (us, gb, de, in…). Defaults to the country a location ends in.
`language`	string	en	Interface language, as a language code with an optional region (en, de, pt-BR, zh-TW…).
`fields`	string	–	Comma-separated projection, e.g. results.title,results.link,knowledge_graph. Cuts tokens.
`format`	string	json	Response format. markdown is LLM-ready prose; compact is JSON with only the fields agents need. Values: json, compact, markdown.

Full reference, error codes and headers are in the /v1/search reference. Queue up to 100 requests at half price with POST /v1/batches.

Use cases

What people build with the Google Knowledge Graph API

Entity grounding for agents

Give an agent a short, sourced summary of a person, company or place before it answers.

Enrichment

Fill in founding dates, headquarters, websites or birth dates for a list of entities.

Brand SERP monitoring

Track what Google's panel says about your brand or executives, and catch changes.

Local entities

Landmarks and businesses come with their type, website and facts such as address or hours.

Pricing