Build AI assistants with all the knowledge in the world

Help your assistant plan, execute, and verify tasks it could not handle from memory alone.

Search that keeps up with the best personal assistants

Answer questions across every subject, at any depth, and send updates before users ask.

Answer everyday questions and specialist research from one API

Give your assistant access to 350M publications, 70M companies, and the live web alongside scores, forecasts, and business hours. Use the same search API for a dinner recommendation, a company lookup, or a scientific question.

Get better answers with fewer tokens

Exa's Dynamic Highlights pull the exact passages that answer each question. Save up to 95% of your input tokens by dropping navigation, ads, and irrelevant content before they get loaded into context.

Update briefings proactively by monitoring in the background

Let your assistant be the first to discover scoops on topics your users care about. Run their recurring questions on a schedule, from a morning news digest to a team's next game or a company they follow. Get notified of updates at your webhook, as a summary or as structured fields.

Exa organizes data across billions of sources

Every datapoint is selected on source quality and freshness, based on the latest events from today

70M+ Businesses
Pricing pages
1B+ People Records
Job postings
Contact data
Product reviews
Latest News
Funding rounds
Financial market data
Exa Connect
Earnings call transcripts
Investor portfolios
Recent hires
Conference speaker lists

Common questions

Exa is the search layer. When a user asks something the model cannot answer from memory, the assistant calls Exa, which searches the live web and returns the relevant pages, passages, and source links. The model writes the answer from what Exa returns and cites the pages.

Exa Search covers sports scores, standings, and schedules for major leagues, weather forecasts, and local business profiles with addresses, hours, and reviews. News becomes searchable within minutes of publication. All of it comes back from the same search call as the rest of the web, with no separate API to integrate for each.

Match the search type to the question. Exa Instant returns results in under 250ms for quick facts like a score. Auto, the default, balances quality and speed in about a second. Deep searches, reads, and synthesizes across many pages in a few seconds for comparisons and research. You choose the type on each request.

Create an Exa Monitor with a search query, a schedule, and a webhook URL. On each run, Exa runs the search, drops results it has already sent, and posts only new ones to your webhook. Each run arrives as a text summary or as JSON that matches a schema you define, with sources attached.

Yes. Highlights return only the passages from each page that answer the query. In Exa's evaluations, 500 characters of highlights matched the accuracy of the first 8,000 characters of page text, and Dynamic Highlights cut tokens by an average of 95% compared to full pages. That keeps the cost of each question down at consumer scale.

No. Exa does not book, buy, send messages, or fill out forms. It supplies the research an assistant needs before it acts, through the API or Exa MCP.

Exa is SOC 2 Type II audited, and honors GDPR and CCPA data-subject rights. Zero data retention is available, meaning your queries and results are not stored or used for training. Single sign-on comes with an Enterprise plan. Terms are on the enterprise page.

Firefox uses Exa to power Smart Window on desktop and Quick Answers on iOS. Firefox chose Exa after evaluating multilingual quality, latency, safety, and privacy.

Give every assistant the context it needs to do great work