Web Data
Web Search API
Up to 30 ranked results per query — merged across engines, deduplicated, with an agreement count — and a deep mode that reads the top pages for you.
/searchfree allowance included30
results per call
1–8
pages read in deep mode
15 min
query cache
multi-engine
agreement signal
About the API
What this endpoint does.
Most search APIs resell a single engine, so every query inherits that engine’s bias and pricing. This one runs its own metasearch cluster: it queries several engines, merges what they agree on, drops known low-value domains, and returns one deduplicated ranked list.
Each result carries an agreement count — how many independent engines surfaced that URL. Roughly one in five results is found by two or more, which makes it a useful relevance hint on top of the ranking. Add deep=1 and the endpoint goes a step further, reading the top pages and returning their extracted text alongside the links, so a research agent gets sources and content in a single call.
Who builds on this
Built for the teams shipping real products.
- 01
AI apps & agents
Ground an answer in fresh sources — deep mode hands back the page text, ready to cite or embed.
- 02
Research & monitoring
Track a topic, brand or competitor with news=1 to see only what broke in the last 24 hours.
- 03
Content & SEO tools
See what already ranks for a query without signing up to a third-party search vendor.
Under the hood
What you send. What comes back.
Request
Inputs you send
qrequiredthe query, 2–400 characters.
limitoptionalresults to return, 1–30. Omitted → 10.
deepoptional1 reads the top pages and returns their text alongside the links.
pagesoptionalhow many pages deep mode reads, 1–8. Omitted → 4.
newsoptional1 restricts results to the last 24 hours.
Response
JSON you get back
resultsranked array of { title, url, snippet, engines, agreement }.
agreementhow many engines returned that URL — a relevance hint, not a guarantee.
engineswhich engines surfaced the result, e.g. ["duckduckgo","bing"].
contentdeep mode only — the readable page body, capped near 5,000 chars.
cachedtrue when the query was answered from the 15-minute cache.
took_msserver-side search time in milliseconds.
Try it
One call, real data.
{
"success": true,
"query": "next.js 16 release",
"count": 10,
"took_ms": 1240,
"deep": false,
"sources_read": 0,
"cached": false,
"results": [
{
"title": "Next.js 16",
"url": "https://nextjs.org/blog/next-16",
"snippet": "Next.js 16 is here with Turbopack by default...",
"engines": [ "duckduckgo", "bing" ],
"agreement": 2
}
]
}Ten results come back — one shown above. Add deep=1 and each result also carries its page text.
Free tier
Ship on this API free, then pay only when your calls outgrow it.
- A free monthly request allowance on your first key — enough to build and test.
- The same key works across every endpoint, so you can switch APIs as you grow.
- When the free requests end, upgrade to paid — your code and endpoints don’t change.
Every endpoint includes a free monthly request allowance — the size differs from API to API. When your free requests run out, you move to a paid plan with the same key. No card needed to start.
Start with Web Search, free.
Create an account and we’ll issue your free API key in minutes — this endpoint included.
© 2026 API Mitra. All rights reserved.
service statusMade in India