SERP API
Get accurate and in real-time results sourced from Google, Bing, and more.
Residential Proxies
Over 100 million real residential IPs from genuine users across 190+ countries.
SCRAPING SOLUTIONS
SERP API
Get accurate and in real-time results sourced from Google, Bing, and more.
Web Scraper API
With 120+ prebuilt and custom scrapers ready for any use case.
Webpage
No blocks, no CAPTCHAs—unlock websites seamlessly at scale.
Scraping Browser
Execute scripts in stealth browsers with full rendering and automation.
PROXY INFRASTRUCTURE
Two integration paths, by project stage
SDK takes a lightweight in-process route; MCP takes a service route—both share the same API key and data source, and can be switched or combined at any time.
SDK
Use it directly in code (best to start)—Wrap Thordata SERP as a LangChain Tool inside the agent process; import and it works.
- Zero deployment, run on install
- Lowest in-process latency
- Instant code changes for easy debugging
- Embeds directly from scripts to notebooks
Typical use: personal projects, demo validation, local and experimental environments.
View integration docsMCP
Serve data as a service (best to scale)—Search runs on the Thordata MCP Server and connects to LangChain through an adapter, fully decoupling tools from agents.
- Independent process that scales horizontally
- Tools versioned and released separately with no agent disruption
- One service reused by many agents and teams
- Choose from stdio / HTTP / SSE
Typical use: production deployment, multi-agent systems, cross-team sharing.
View integration docsIntegration steps
Three steps from API token to a working, real-time-search-capable LangChain agent.
1
Get an API token
Create a Thordata account for a free API token, no credit card.
2
Choose an integration path
Use the SDK to wrap a Tool for custom development; choose MCP to connect existing systems or multi-agent workflows.
3
Start getting real-time results
The agent decides when to search and returns structured data ready to use.
STEP 1Install SDK (BASH)
pip install langchain-thordata
STEP 2Connect Agent (PYTHON)
import os
os.environ["THORDATA_API_KEY"] = "your-api-key"
from langchain_thordata import ThorDataSearchTool
from langchain_openai import ChatOpenAI
# 1. Initialize the search tool
tool = ThorDataSearchTool()
# 2. Bind it to your model. The agent decides when to run live web search
llm = ChatOpenAI(model="gpt-4o-mini", temperature=0)
model_with_tools = llm.bind_tools([tool])
# 3. Ask a question. The model triggers real-time search when needed
response = model_with_tools.invoke("Search for coffee")
print(response)Real Return Example - JSON
1{2"engine": "google",3"query": "coffee",4"taskId": "65f2a1b8c9d0e4f7a1b2",5"result": {6 "organic": [7 {8 "link": "https://en.wikipedia.org/wiki/Coffee",9 "display_link": "en.wikipedia.org",10 "title": "Coffee",11 "description": "Coffee is a beverage brewed from roasted beans."12 },13 {14 "link": "https://www.britannica.com/topic/coffee",15 "display_link": "britannica.com",16 "title": "Coffee | Origin, Types, Uses, History, & Facts",17 "description": "Coffee, beverage brewed from the roasted seeds of the fruit."18 }19 ],20 "people_also_ask": [21 {22 "question": "How is coffee made?",23 "answers": "Coffee is brewed from roasted coffee beans."24 }25 ]26}
SDK and MCP are both standard LangChain Tools. They work with Calculator, Browser, and other tools and return the same structure. The SDK also includes history and statistics tools for troubleshooting and usage monitoring.
Why choose the Thordata LangChain integration?
Keep LangChain agents supplied with real-time, reliable, structured search data.
- Multi-engine integration—One API call to Google, Bing, and DuckDuckGo, with multilingual search across 190+ countries.
- Agents that search on demand—The LangChain agent calls the search tool itself when it needs live data, with no manual orchestration.
- Optimized token usage—Results are cleaned and organized automatically, keeping only the most relevant content to cut token use and speed up responses.
- Enterprise reliability—Stable API performance, automatic retries, and developer support keep AI agents reliable at scale.
Agent-native development—more than a search API, a complete agent development stack.
- Fast integration—Register a LangChain tool in a few lines of code.
- MCP support—Search as a standalone service, connected to many agents and workflows.
- Easy debugging—Stable structured response fields make issues easy to locate.
- Fully managed infrastructure—100M+ residential IPs, CAPTCHA handling, and retries backed by the platform.
Core capabilities
A capability set designed for LangChain agents—retrieval, targeting, output, and scale in one.
Real-time SERP data
Live Google, Bing, and DuckDuckGo results, so agent answers are always based on the current web.
190+ geo-targeting
Fetch by country, city, language, and device—see what your target-market users see.
Structured output
Stable fields, consumed directly by agents with no cleaning.
Managed collection layer
100M+ rotating residential IPs, JavaScript rendering, and CAPTCHA handling, all covered by the platform.
Fine-grained retrieval control
Tune pagination, search type, domain, and device to scope data precisely.
Smooth scaling
From prototype to enterprise on the same key, no rebuild, transparent pay-as-you-go.
Typical use cases
See how Thordata helps developers and teams build AI apps with real-time search.
Real-time information assistant
Let the agent retrieve and answer with the latest web information, always current.
Research and analysis assistant
Automatically gather and merge multi-source information to support research and competitive analysis.
Real-time Q&A system
Give accurate answers from the latest data—prices, rankings, product launches.
Content generation assistant
Draft articles, summaries, and research reports from live web information.
Frequently asked questions
What is the difference between the LangChain integration and calling the API directly?
The integration adds an advanced abstraction layer built for AI agents: automatic tool calls, task orchestration, and seamless fit into agent workflows, with less manual request handling. You can still call the API directly for full control over request structure and custom pipelines.
How do I choose between SDK and MCP integration?
Choose the SDK for fast prototyping and single-agent apps—in-process, zero deployment, low latency. Choose MCP for production and multi-agent systems—a standalone service that scales, is shared across agents, and versioned independently.
Why choose Thordata over other SERP APIs?
190+ country geo-targeting, fully managed proxies and CAPTCHA, structured JSON / Markdown output, transparent pricing from $0.70 per 1K, and 500 free responses on signup—test first, pay later.
Which search engines and sources are supported?
Google, Bing, and DuckDuckGo, with geo-targeting across 190+ countries and cities, plus multilingual and desktop / mobile device parameters.
How does the free quota work, and what happens after it?
Sign up for 500 free SERP responses over 7 days, no credit card. Beyond that, pay as you go from $0.70 per 1K, with no upfront cost.
Ready to build smarter AI agents?
Start with the Thordata SERP API for free and add real-time search to your LangChain agents.