
by Exa
Exa 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.
APIUsage-based
Step 1: Get an API key from the Exa dashboard.
Step 2: Install the Python or JavaScript SDK, or call the API with cURL.
Step 3: Run a search and request highlights or full contents.
Step 4: Use the agent endpoint with an output schema for structured research results.
Step 5: Hand search tools to an LLM, or connect through the MCP server.
Step 6: Set up monitors to run searches on a schedule.
You pass an output schema with a deep search or an agent run, and the result comes back as structured fields with citations.
The pricing page lists web, people, company and scholarly works indexes, and the home page lists news and financials categories.
Zero data retention is available on the enterprise plan on supported products, so queries and results are not stored or trained on.
They are third-party data sources reachable through one Exa integration, billed on top of Agent usage. Some are self-serve and others are available on request.
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.