Real-time search
Search returns real-time web search data and is positioned for latency-sensitive agent workflows, with configurable latency from 180ms to 1s.
Exa is a web search API for AI apps needing realtime web data, structured results, and grounded citations. Supports search, research, retrieval, monitoring, and enrichment.
Exa is a web search API and AI search platform built for applications that need current information from the web. The site describes it as a modern AI search engine with search, crawling, deep research, and monitoring products for agent workflows.
Its core value is helping AI systems retrieve realtime web data, structured results, and grounded citations for tasks such as coding assistance, research, company and people lookup, enrichment, and news monitoring. Pricing information also shows a free tier and paid usage across multiple endpoints, with enterprise options for higher-volume or custom deployments.
Search returns real-time web search data and is positioned for latency-sensitive agent workflows, with configurable latency from 180ms to 1s.
Deep is built for complex search tasks and multi-step research, with structured outputs and web-grounded citations.
Contents provides full-page web contents with token-efficient highlights and configurable livecrawl policies for AI context.
Monitors tracks new events and updates across the web on a specified cadence and can send updates through webhooks.
Agent supports asynchronous research, list building, and enrichment with structured outputs and citations.
The pricing page shows enterprise options for custom datasets, custom rate limits, moderation, zero data retention, and onboarding support.
Build coding assistants that search docs, repositories, and changelogs in real time so answers reflect current package and API references.
Support lead enrichment or company research workflows with structured company data and grounded citations.
Run complex research queries that require structured outputs, such as product, market, or scientific lookup tasks.
Track fresh web events and updates on a schedule, then push results through webhooks or downstream automations.
Retrieve full-page web content and highlights to give models more context while keeping token use efficient.
Exa is a web search API for AI applications. The source pages describe Search for real-time search results, Deep for multi-step research workflows, Contents for full-page web content with highlights, and Monitors for tracking new web events on a cadence.
The pricing page shows a Free tier with up to 20,000 requests per month, plus paid usage for Search, Deep Search, Contents, Monitors, and Agent. The site also offers enterprise and startup/education options.
The source pages show Exa being used for coding agents, chatbots, news monitoring, enrichments, voice agents, people search, and company search. The product pages also emphasize document, repo, and company-data retrieval workflows.
Search returns standard search results with URLs and page contents, plus AI summary and structured outputs. Deep is positioned for complex search tasks with structured outputs and web-grounded citations.