
by Tavily
Tavily is a web access API for AI agents, covering search, content extraction, crawling and research. It returns structured, chunked results so models can answer from fresh web data.
APIUsage-based
Step 1: Sign up for a free account and copy your API key from the dashboard.
Step 2: Install the Python or JavaScript SDK, or call the HTTP endpoint with a Bearer token.
Step 3: Call search with a query and options such as search depth and max results.
Step 4: Use extract to pull content from specific URLs, or research for a multi-source report.
Step 5: Pass the returned chunks to your model as grounding context.
Requests stop until your credits reset on the first day of the month or you upgrade to a paid plan.
The extract endpoint accepts one URL or a batch of up to 20. Failed URLs appear in a separate failed_results array, so check both arrays even when the response is HTTP 200.
Yes. Requests pass through security, privacy and content validation layers that block PII leakage, prompt injection and malicious sources.
The docs show a Python SDK, a JavaScript SDK and examples for cURL, PHP, Go, Java and Ruby against the HTTP API.
ExaExa is a search API for AI agents that returns web data, page contents and structured research results. It offers search, contents, agent and monitor endpoints.
FirecrawlFirecrawl is a web data API that searches, scrapes and interacts with web pages and returns clean, LLM-ready data. It is open source and works with AI agents through MCP and a CLI.