Search API for AI agents over a huge web index, with token-efficient highlights and research endpoints.
Exa homepage
1 / 2Overview
What is Exa 3.0?
Smarter search that gives AI agent richer data at lower cost.
Exa 3.0 is the latest search-quality update to Exa's search API for AI agents, adding a larger embedding model, billions of new documents (including 1.27B ecommerce listings and 211M+ local businesses) and Dynamic Highlights that use 95% fewer tokens than full page text.
The API offers configurable latency tiers (Lite 450ms, Auto 1s, Deep ~10s) across verticals such as Full Web, News, Companies, Research, People and Financials. It returns results, answers or structured outputs with grounded citations, while the Contents API retrieves full page contents for LLM context and Monitors run scheduled searches with webhook updates. Enterprise-grade controls include SOC 2 Type II, GDPR/CCPA handling, HIPAA support and Zero Data Retention.
Pricing starts free: the Starter tier gives $10 in credits every month plus a $10 onboarding bonus with no payment method required. A new Exa Lite tier costs $4 per 1,000 requests and Exa Auto costs $7; Enterprise plans are custom.
Platforms and languages
Exa 3.0 Availability
Platforms
Languages
Capabilities
Exa 3.0 Key Features
Configurable latency tiers
Search API offers Lite (450ms), Auto (1s) and Deep (~10s) tiers across Full Web, News, Companies, Research, People and Financials verticals.
Token-efficient Highlights
A specialized model extracts the most relevant excerpts, cutting token usage by about 90-95% versus full-page text.
Grounded results and answers
Returns results, answers or structured outputs with grounded citations, scaling from low-latency search to deep research runs.
Contents API and Monitors
The Contents API retrieves full page contents for LLM context with livecrawl policies, while Monitors run scheduled searches with webhook updates.
Enterprise-grade controls
SOC 2 Type II, GDPR/CCPA handling, HIPAA support with a BAA, and Zero Data Retention options.
Best for
Who uses Exa 3.0?
Developers and AI engineering teams building AI assistants, coding agents and research tools
Add web search, content extraction and grounded answers to AI applications via API.
Plans and access
Exa 3.0 Pricing
Freemium
Free Starter ($10 credits/mo, no card); Lite $4 / Auto $7 per 1k requests; Enterprise custom.
Common questions
Exa 3.0 FAQs
What is Exa?
A web search and research infrastructure for AI applications: its own search index plus content extraction and agentic research APIs so apps can find sources, extract content and generate grounded output.
How large is Exa's index?
As of October 2026 the index tracks trillions of URLs with 100 billion pages from the public web, and it changes continuously as pages are discovered, refreshed and removed.
What data-retention options exist?
Production and enterprise setups get Zero Data Retention (queries and results are not stored or used for training), and HIPAA compliance is available for eligible enterprise customers.
How does pricing work?
API usage is deducted from account credits based on the endpoint and options used. New accounts receive free credits; paid usage is pay-as-you-go except for enterprise contracts.
Reviews
See what the community thinks and share your experience.
—
Based on 0 ratings
Rating distribution
Community reviews
No reviews
No written reviews yet.
