-
Under 1-second latency
Premium infrastructure delivering search results faster than ever for time-sensitive applications. -
Perfect for AI workflows
Seamless integration with LangChain, LlamaIndex, and custom AI frameworks. -
Real-time enrichment
Enable multi-step reasoning and dynamic information retrieval for intelligent agents.
Web Search API
Access search results from 7 major search engines through a unified API. Power AI agents, SEO platforms, and business intelligence tools with real-time search data.
- Pay only for successful delivery
- Instant search results, under 1 second
- Data delivered in JSON, HTML or Markdown
- Geo-location targeting - FREE
Trusted by 20,000+ customers worldwide
Query Multiple Search Engines Through One API
Access search data from global and regional search engines covering 195 countries with consistent JSON, HTML, or Markdown output formats
Global Search Engines
Access worldwide search engines with comprehensive coverage of organic results, paid ads, knowledge graphs, and rich search features.
Regional Leaders
Query dominant regional search engines in Russia, China, Korea, and other key markets with native language support.
Privacy-Focused Options
Access unbiased, non-personalized search results from privacy-first engines for competitive analysis and unfiltered research.
import requests
url = 'https://api.brightdata.com/request'
headers = {
'Authorization': 'Bearer ',
'Content-Type': 'application/json'
}
payload = {
'zone': 'web_search',
'url': 'https://www.google.com/search?q=AI+trends+2026',
'format': 'json' # Get structured JSON output
}
response = requests.post(url, json=payload, headers=headers)
data = response.json()
# Access organic results
for result in data.get('organic', []):
print(f"{result['title']}: {result['link']}")
const response = await fetch('https://api.brightdata.com/request', {
method: 'POST',
headers: {
'Authorization': 'Bearer ',
'Content-Type': 'application/json'
},
body: JSON.stringify({
zone: 'web_search',
url: 'https://www.google.com/search?q=AI+trends+2026',
format: 'json'
})
});
const data = await response.json();
// Access organic results
data.organic?.forEach(result => {
console.log(`${result.title}: ${result.link}`);
});
curl -X POST 'https://api.brightdata.com/request'
-H 'Authorization: Bearer '
-H 'Content-Type: application/json'
-d '{
"zone": "web_search",
"url": "https://www.google.com/search?q=AI+trends+2026",
"format": "json"
}'
# LangChain integration example
from langchain.tools import Tool
import requests
def web_search(query: str) -> str:
"""Search the web using Web Search API"""
response = requests.post(
'https://api.brightdata.com/request',
headers={'Authorization': 'Bearer '},
json={
'zone': 'web_search',
'url': f'https://www.google.com/search?q={query}',
'data_format': 'markdown' # Clean Markdown for LLMs
}
)
return response.text
# Create LangChain tool
search_tool = Tool(
name="WebSearch",
func=web_search,
description="Search the web for current information"
)
Built for AI Agents, SEO Platforms, and Enterprise Intelligence
Trusted by leading technology companies worldwide
AI Agents & RAG
Power AI agents with real-time web search for data enrichment, fact-checking, and retrieval-augmented generation workflows
SEO Platforms
Build rank tracking tools and monitoring dashboards with localized search data across engines and locations
Market Intelligence
Analyze search trends, competitor strategies, and market opportunities across industries and regions
Brand Protection
Monitor brand mentions and sentiment across search results globally to protect reputation
High-Performance Infrastructure for Scale
-
No proxy management
72M+ residential IPs automatically rotated for every request—no blocks or bans. -
CAPTCHA-free operation
Automatic CAPTCHA solving and fingerprinting handled transparently. -
Focus on data, not infrastructure
Spend time building applications, not maintaining scraping infrastructure.
How Web Search API works
Web Search API Pricing
About Bright Data's Web Search API
Global coverage
See search results as customers do in all 195 countries
Under 1-second delivery
Get fresh search results data faster than your users can refresh the page
Flexible, structured data
Clean, structured JSON, HTML or Markdown, no parsing headaches
Enterprise-grade performance
99.9% uptime SLA, get the search data you need worldwide instantly
Never pay for failed requests
Charges you only for successful data retrieval
Predictable pricing
No maintenance headaches, no wasted budget on failed attempts
Trusted by LeadingTechnology Companies
- 99.9% uptime guarantee
- 195 countries coverage
- Sub-second responses
- Enterprise-grade security
Enterprise Scale
Handle millions of daily requests with consistent performance
GDPR Compliant
Fully compliant with data privacy regulations worldwide
Expert Support
Dedicated account managers and 24/7 technical assistance
Complete Developer Experience
Step-by-step guides and code examples in every language
Official libraries for Python, Node.js, and more
Dashboard with performance metrics and usage analytics
Works with LangChain, LlamaIndex, and custom frameworks
Tailored implementations for enterprise requirements
Powering AI and Enterprise Applications
Start building with Web Search API
Web Search API FAQs
What is a Web Search API?
A Web Search API (also known as AI Search API) provides programmatic access to search engine results through a unified endpoint. Instead of manually querying search engines or managing complex scraping infrastructure, developers can retrieve structured search data via simple API calls for AI applications, SEO platforms, business intelligence, and research tools. Our API supports 7 major search engines with consistent data formats.
Which search engines does the Web Search API support?
We support 7 major search engines covering 195 countries through a single unified API:
- Global search leaders with worldwide coverage
- Regional dominators in Russia and CIS countries
- China's #1 search platform
- Korea's leading search engine
- Privacy-focused search options
- Popular engines in Japan and Taiwan
All with consistent JSON/HTML/Markdown output, geo-targeting, and pay-only-for-success pricing.
How is this different from other search APIs?
Unlike single-engine APIs, our Web Search API provides unified access to multiple search engines through one endpoint with consistent data formats. We handle all proxy management, CAPTCHA solving, and anti-bot bypassing automatically. You get sub-second responses optimized for AI agents, automatic retry logic with 72M+ residential IPs, city-level geo-targeting across 195 countries, and only pay for successful requests.
Can I use this for AI agents and LLMs?
Absolutely. Our Web Search API is optimized for AI applications including:
- RAG (Retrieval-Augmented Generation): Real-time context enrichment
- AI Agents: Sub-second responses for multi-step reasoning workflows
- LLM Integration: Markdown output format perfect for prompt context
- Fact-Checking: Cross-reference information across multiple engines
Seamless integration with LangChain, LlamaIndex, and custom AI frameworks.
What output formats are available and when should I use each?
Web Search API offers three output formats:
- JSON: Structured data perfect for applications, databases, and analytics workflows
- Markdown: Clean, readable format ideal for LLMs, AI agents, and content analysis
- HTML (raw): Complete page data for custom parsing or archival purposes
Most developers use JSON for programmatic access, while AI/LLM applications prefer Markdown for cleaner context.
How fast are the API responses?
Standard requests complete in under 5 seconds. For AI agents and real-time applications requiring sub-second performance, optimized routing infrastructure delivers complete search results in under 1 second. Contact sales for access to high-performance endpoints.
How does geo-targeting work?
Web Search API supports 195 countries with city-level targeting precision. Specify country codes, regions, cities, or coordinates in your request. We route queries through real residential IPs from our 72M+ IP pool in target locations, ensuring authentic results as seen by actual users. Perfect for local SEO, market research, and geo-specific analysis.
What's included in the pricing?
Pay-per-request pricing with pay only for success:
- All 7 search engines included (no per-engine surcharges)
- All output formats (JSON, Markdown, HTML)
- Geo-targeting in 195 countries
- Automatic unlocking (proxy rotation, CAPTCHA solving)
- Volume-based discounts
You're only charged when search results are successfully delivered.
Is there a limit to the number of concurrent requests?
No hard limits for enterprise accounts. Our infrastructure scales to handle millions of requests daily. Standard accounts support multiple concurrent requests. For applications requiring thousands of concurrent searches or custom limits, contact sales for dedicated infrastructure.
When should I use asynchronous mode instead of regular requests?
Use async mode for high-volume batch operations (1,000+ queries), non-urgent data collection, or when you need maximum reliability. Async offers:
- 99.99% success rate (higher than sync)
- Cost savings: only the "send request" is billed; collecting responses is completely free
- Responses stored for 48 hours - retrieve multiple times at no cost
- Usually ready in 5 minutes (max 8 hours during peak)
Perfect for overnight batch jobs, competitor monitoring, or large-scale keyword research.
How does async mode billing work?
You're only charged when sending the request. Collecting the response is free, and you can retrieve it multiple times within 48 hours at no additional cost. This means if you need to download the same results multiple times, or if your first download fails, you won't be charged again.
How do you handle CAPTCHAs and anti-bot measures?
All anti-bot mechanisms are handled automatically:
- CAPTCHA Solving: Automatic resolution without manual intervention
- Proxy Rotation: Each request uses different IPs from 72M+ residential pool
- Fingerprinting: Browser fingerprint management and header optimization
- Retry Logic: Automatic retries with different IPs on failures
You make API calls—we handle all unlocking challenges transparently.