Playground Sign in Start free
SERP API

SERP API for Google search results, as clean JSON.

Scrape Google search results in real time: web results with People also ask and Google AI Overviews, plus news, images, videos, Maps places, jobs and Google AI Mode. 5 credits per request, any type. Blocked or failed requests cost nothing.

5,000 free credits (1,000 searches) for 7 days. No card.

GET /serp · 5 credits
curl -G https://api.ujeebu.com/serp -H "ApiKey: YOUR_API_KEY" --data-urlencode 'search=beste projektmanagement software' --data-urlencode 'location=de' --data-urlencode 'lang=de' --data-urlencode 'device=mobile'
{
  "organic_results": [{
    "position": 1, "site_name": "Asana", "cite": "https://asana.com",
    "title": "Projektmanagement-Tools: Die 14 besten im Vergleich [2026]",
    "date": "27.05.2026", "description": "1. Asana: Die beste Wahl für ..." }, ...],
  "related_questions": ["Welches Programm für Projektmanagement?", ...],
  "ai_overview": { "text": "Die Wahl der besten Projektmanagement-Software ...",
    "text_blocks": [...], "references": [8 sources] }
}
5credits per search, any result type
7Google result types from one endpoint
100results in one request with batch
99.6%of our web search health probes succeeded (525 of 527, Oct 10 to Oct 21, 2026)
0credits for blocked or failed requests
Search types

Every Google result type, one API

One endpoint, /serp, and a search_type parameter. Every type takes the same location, language and device options and costs the same 5 credits. The samples below are trimmed from real responses of October 21, 2026.

Web · search_type=search

Google search results API: organic results, People also ask and Google AI Overviews

The default search returns the Google results page as JSON: every organic result with its position, plus the People also ask questions and the Google AI Overviews block when Google shows them. It is the rank tracking API behind SERP monitoring tools, by country, city, language and device.

What you get
organic_results
position, title, link, description, cite, site_name, date
related_questions
the People also ask questions
ai_overview
Google AI Overviews: text, text_blocks, references (null when Google shows none)
top_stories, knowledge_graph, videos
when Google shows them
pagination, metadata
next pages, the Google URL, number of results
Built for
  • Rank tracking and SEO tools. Record the position of every domain for your keywords, per country, city and device, and chart it over time.
  • AI agents and RAG. Give an agent a search tool that returns fresh results with sources, through the API or the Ujeebu MCP server.
  • Content research. Collect the People also ask questions and the sources Google AI Overviews cite for your topics.
Request
curl -G https://api.ujeebu.com/serp -H "ApiKey: YOUR_API_KEY" --data-urlencode 'search=beste projektmanagement software' --data-urlencode 'location=de' --data-urlencode 'lang=de' --data-urlencode 'device=mobile'
Response (trimmed) · 200 OK · 5 credits
{
  "organic_results": [{
    "position": 1, "site_name": "Asana", "cite": "https://asana.com",
    "title": "Projektmanagement-Tools: Die 14 besten im Vergleich [2026]",
    "date": "27.05.2026", "description": "1. Asana: Die beste Wahl für ..." }, ...],
  "related_questions": ["Welches Programm für Projektmanagement?", ...],
  "ai_overview": { "text": "Die Wahl der besten Projektmanagement-Software ...",
    "text_blocks": [...], "references": [8 sources] }
}

People also ask API

related_questions lists the People also ask questions Google shows for the query, in order: a ready list of the questions searchers ask about a topic.

Google AI Overviews API

ai_overview holds the Google AI Overviews answer (text, text_blocks) and the sources it cites (references) in the same 5-credit call. It is null when Google shows no overview for the query.

News · search_type=news

Google News API

search_type=news returns the Google News results for a query: headline, publisher, how long ago it was published, the summary and the thumbnail, from the country and language you choose.

What you get
news
position, title, link, siteName, date (as Google shows it, e.g. "40 minutes ago"), description, image, favicon
pagination, metadata
next pages, the Google URL, number of results
Built for
  • Media monitoring. Search your brand, competitors or tickers on your own schedule and store each new headline with its publisher.
  • Market research. Follow how a topic is covered across countries and languages, one search per market.
  • AI agents and RAG. Ground answers about current events in recent headlines, with the publisher of each one.
Request
curl -G https://api.ujeebu.com/serp -H "ApiKey: YOUR_API_KEY" --data-urlencode 'search=electric vehicle tariffs' --data-urlencode 'search_type=news'
Response (trimmed) · 200 OK · 5 credits
{
  "news": [{
    "position": 1, "siteName": "Automotive News", "date": "40 minutes ago",
    "title": "UK open to discussing Chinese car tariffs ahead of EU summit",
    "description": "Business Secretary Jonathan Reynolds signaled he is open to ...",
    "image": "data:image/jpeg;base64,..." }, ...],
  "metadata": { "number_of_results": 5580, "query_displayed": "electric vehicle tariffs" }
}
Images · search_type=images

Google Images API

search_type=images is a Google image search API: the full-size image URL, Google's thumbnail, the page the image is on and its size, plus the refinement chips Google suggests. Our test search returned 100 images in one call.

What you get
images
position, title, image (full size), google_thumbnail, link (page), source, domain, width, height
suggestions
related searches Google offers as chips: title, thumbnail, api_link
Built for
  • Product and price intelligence. See which retailers and product photos show up for a product search, with the page each image links to.
  • Datasets and catalog enrichment. Collect candidate images with their size and source page, then check usage rights at the source.
  • Brand monitoring. Find where your logo or product images appear for the searches that matter to you.
Request
curl -G https://api.ujeebu.com/serp -H "ApiKey: YOUR_API_KEY" --data-urlencode 'search=standing desk' --data-urlencode 'search_type=images'
Response (trimmed) · 200 OK · 5 credits
{
  "images": [{
    "position": 1, "source": "Varidesk", "domain": "www.vari.com",
    "title": "VariDesk Pro Plus 36 in. Height-Adjustable Standing Desk Converter",
    "image": "https://www.vari.com/dw/image/v2/...", "width": 800, "height": 800,
    "google_thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=...",
    "link": "https://www.vari.com/sit-stand-converter-varidesk-pro-plus-36/DC-PP36.html" }, ...],
  "suggestions": [{ "position": 1, "title": "Office", "api_link": "..." }, ...]
}
Videos · search_type=videos

Google Videos API

search_type=videos returns Google's video results: YouTube and the other video sites Google indexes, with the channel, duration and publish date of each video.

What you get
videos
position, title, url, author, provider, duration, pubDate, description
pagination, metadata
next pages, the Google URL, number of results
Built for
  • Video SEO. Track which videos and channels rank in Google for your keywords, and how long they are.
  • Creator and brand research. List the creators who cover a topic or a product, across YouTube and the other video sites.
  • AI agents and RAG. Let an assistant suggest current videos on a topic, with the channel and date of each.
Request
curl -G https://api.ujeebu.com/serp -H "ApiKey: YOUR_API_KEY" --data-urlencode 'search=pour over coffee tutorial' --data-urlencode 'search_type=videos'
Response (trimmed) · 200 OK · 5 credits
{
  "videos": [{
    "position": 1, "title": "A beginner's guide to POUR-OVER COFFEE",
    "author": "Majordomo", "provider": "YouTube",
    "duration": "6:07", "pubDate": "Dec 8, 2023",
    "description": "Our friend Alex Kaplan (once the youngest certified Q Coffee Grader ..." }, ...],
  "metadata": { "number_of_results": 2180000 }
}
Maps · search_type=maps

Google Maps scraper API: places as JSON

search_type=maps returns the places Google Maps lists for a search, as a local searcher sees them: put a city in location and get each business's address, phone, website, rating, review count, hours and coordinates. A simple way to scrape data from Google Maps without running a browser. Our test returned 20 places.

What you get
maps_results
position, title, category, address, phone, website, rating, reviews, opening_hours, price, gps_coordinates, place_id, cid, thumbnail
pagination, metadata
next pages, the Google URL
Built for
  • Local business intelligence. Map competitors by category and city, with ratings, review counts and price levels.
  • Local SEO. Track where a business ranks in Google Maps for each search and town; place_id and cid keep the same business joined across runs.
  • Lead lists. Build lists of local businesses with phone and website, ready for your CRM.
Request
curl -G https://api.ujeebu.com/serp -H "ApiKey: YOUR_API_KEY" --data-urlencode 'search=coffee shops' --data-urlencode 'search_type=maps' --data-urlencode 'location=Austin, TX'
Response (trimmed) · 200 OK · 5 credits
{
  "maps_results": [{
    "position": 1, "title": "Bennu Coffee on Congress", "category": "Cafe",
    "address": "515 S Congress Ave, Austin, TX 78704", "phone": "(512) 448-3919",
    "website": "http://www.bennucoffee.com/", "rating": 4.5, "reviews": 1610,
    "opening_hours": "Open 24 hours", "price": "$1–10",
    "gps_coordinates": { "latitude": 30.256083, "longitude": -97.74623 },
    "place_id": "ChIJ7agecQG1RIYRmhih01aU_-A", "cid": "16212840284195723418" }, ...]
}
Jobs · search_type=jobs

Google Jobs API

search_type=jobs returns Google Jobs listings with the company, place, salary when posted, highlights and every site you can apply on. Filter by place and radius, remote, date posted and employment type, and fetch one listing again later with its job_id.

What you get
jobs
position, title, company_name, location, via, description, job_highlights, detected_extensions (posted_at, schedule_type, salary), extensions, apply_options, share_link, thumbnail, job_id
job (search_type=job)
one listing again by its job_id, or a 404 when it is gone
Built for
  • Job boards and aggregators. Fill a board with fresh listings by role and city; job_id keeps duplicates out.
  • HR tech and labor market analytics. Count openings by employer, role and city, and collect posted salary ranges.
  • Recruiting. See who is hiring for the roles you place and where each job is posted.
Request
curl -G https://api.ujeebu.com/serp -H "ApiKey: YOUR_API_KEY" --data-urlencode 'search=data engineer' --data-urlencode 'search_type=jobs' --data-urlencode 'location=Austin, TX' --data-urlencode 'date_posted=week'
Response (trimmed) · 200 OK · 5 credits
{
  "jobs": [{
    "position": 1, "title": "Data Engineer (Intermediate OR Senior)",
    "company_name": "Teacher Retirement System of Texas", "location": "Austin, TX",
    "via": "Teal", "job_id": "AsxPpyECmDhA6w7hAAAAAA==",
    "detected_extensions": { "posted_at": "2 days ago", "schedule_type": "Full-time" },
    "apply_options": [{ "title": "Teal", "link": "https://www.tealhq.com/job/..." }, ...],
    "job_highlights": [...], "description": "About the position ..." }, ...]
}
AI Mode · search_type=ai

Google AI Mode API

search_type=ai returns the Google AI Mode answer to a question, as text and blocks, with the sources it cites. An AI Mode answer can take up to a minute.

What you get
ai_results
answer_text, text_blocks (paragraphs and lists), citations (title, link, source, description), is_complete
Built for
  • AI visibility tracking. See which brands Google AI Mode recommends for your buyers' questions and which sites it cites.
  • Content and PR research. Find the pages AI answers lean on for your topics.
  • AI agents and RAG. Compare your own answers with Google's, with the citations side by side.
Request
curl -G https://api.ujeebu.com/serp -H "ApiKey: YOUR_API_KEY" --data-urlencode 'search=what is the best crm for a small business' --data-urlencode 'search_type=ai'
Response (trimmed) · 200 OK · 5 credits
{
  "ai_results": {
    "answer_text": "The best CRM for a small business depends entirely on your specific ...",
    "text_blocks": [{ "type": "paragraph", "content": "..." }, { "type": "list", ... }],
    "citations": [{ "source": "PCMag",
      "title": "The Best Small Business CRM Software We've Tested for 2026 | PCMag",
      "description": "The approachable, flexible, and low-cost Bigin by Zoho CRM ..." }, ...]
  }
}
See it in action: the AI Citation Index Google AI Overviews come back in the same SERP call. Every week we publish which brands they recommend for real buyer questions, and the sites they cite.
Open the index
Options

Request options: location, language, device and more

Target a country or a city, a language and a device on every search type. The full list is in the SERP API docs.

ParameterWorks withValuesNotes
searchall typesthe queryOr url: a full Google search URL.
search_typeall typessearch, news, images, videos, maps, jobs, job, aiDefault search (web).
locationall typesa 2-letter country code (us, de) or a place name (Austin, TX)Default US, up to 100 characters.
langall typesISO 639-1 code (en, de, fr)Default en. With a place outside English-speaking countries, use the local language.
deviceall typesdesktop, mobileDefault desktop.
results_countall types1 to 100Without batch, one page (about 10 results).
pageweb, news, images, videos, maps, jobs1 to 40Each response also carries the next page URLs. Not for the job lookup.
batchweb search and jobstrueUp to 100 results in one request, numbered without duplicates. 5 credits per 10 results.
radiusjobs1 to 1000 (km)Around the place in location.
remotejobstrueRemote listings only.
date_postedjobstoday, 3days, week, monthListings posted in that window.
employment_typejobsFULLTIME, PARTTIME, CONTRACTOR, INTERN, TEMPORARYComma-separated = any of them.
job_idjob (lookup)a job_id from a jobs resultFetches that listing again; send search = the listing's title, company and location (a bare keyword answers 404). 404 also when it is gone.
extra_paramsall typesother Google URL parameters, e.g. &safe=activeFor Google options without their own parameter.
with_html, html_onlyall typestrueAdd the raw HTML of the results page, or get only the HTML.

More than 10 results, per type

Webpage 1 to 40, or batch=true with results_count up to 100 in one request
News, Images, Videos, Mapspage 1 to 40
Jobspage 1 to 40 (10 jobs per page), or batch=true with results_count up to 100 in one request
Job lookup, AI Modeone listing or one answer per request
Pricing

SERP API pricing: 5 credits a search

Every search type costs the same. A batch costs 5 credits per 10 results requested (results_count=30 is 15 credits). Blocked or failed requests cost nothing, and the plans share one credit pool with every Ujeebu API. See full pricing →

Start free

Free trial: 5,000 credits (1,000 searches) for 7 days, no card. No subscription? Top-up packs: from $12 for 50K credits ($0.24 per 1K) down to $0.13 per 1K at 5M credits. Top-up credits last 60 days under 500K credits, 90 days from 500K and 180 days from 5M.

Compare

A SerpApi and Serper alternative, compared

Entry plans as listed on our comparison pages (October 2026). Prices change: each page has the details and the sources.

Questions

SERP API questions

What is a SERP API?

A SERP API runs a search on a search engine and returns the results page (the SERP) as structured data instead of HTML. Ujeebu's SERP API covers Google: web results with People also ask and Google AI Overviews, plus news, images, videos, Maps places, jobs and Google AI Mode, all as JSON from one endpoint with the search_type parameter.

How much does the SERP API cost?

Every request costs 5 credits, whatever the search type, and a batch costs 5 credits per 10 results requested. The Starter plan ($49 a month, 300,000 credits) covers 60,000 searches. Failed requests (blocks, timeouts, server errors) cost nothing, and the Ujb-Credits response header gives the credits each call used.

Is there a free SERP API plan?

Every new account gets 5,000 free credits for 7 days, no card needed: that is 1,000 searches to test every search type. After that, pick a monthly plan or a one-time top-up pack.

Can I scrape Google search results for a specific city or language?

Yes. Set location to a country code (us, de) or a place name such as "Austin, TX", lang to the language, and device to desktop or mobile. Each request fetches the results page from Google for that place, language and device.

Is it an alternative to the Google Custom Search API?

It answers a different need. The Custom Search JSON API searches a Programmable Search Engine you configure in Google Cloud. Ujeebu returns the Google results page a searcher sees, with People also ask, Google AI Overviews and the news, images, videos, Maps and jobs results, with one API key and no search engine to set up.

Does it return Google AI Overviews and Google AI Mode?

Yes. A web search returns Google AI Overviews in the ai_overview field when Google shows one (null otherwise), and search_type=ai returns the Google AI Mode answer with its citations. Both cost 5 credits.

How do I get more than 10 results?

Use page (1 to 40) to walk the pages of web, news, images, videos, maps and Google Jobs results; each response carries the next page URLs. For web search and Google Jobs, batch=true with results_count up to 100 returns them in one request, billed 5 credits per 10 results.

How does Ujeebu compare with SerpApi and Serper?

Ujeebu covers Google only, and SERP shares one credit pool with our scraping, extraction, screenshot and Markdown APIs. Our comparison pages list each provider's plans, prices and features side by side: Ujeebu vs SerpApi, Ujeebu vs Serper and SerpApi vs Serper.

Your first Google results in one request.

5,000 free credits, 1,000 searches, no card. Every search type, the same 5 credits.