Exa is web search and data infrastructure built for AI. A single API lets your applications and agents find the exact content they need on the web and turn it into clean, structured, up-to-date data.
Exa gives your applications and agents a single API to find the exact content they need on the web and turn it into clean, structured, up-to-date data. Core capabilities include:
Search: find the most relevant webpages using Exa's embeddings-based and keyword search.
Contents: retrieve clean, parsed, token-efficient content from any page or search result.
Research: run deep, multi-step research that returns grounded answers with citations and structured outputs.
Exa continually expands its capabilities across search, crawling, and research. See the documentation for the latest features and endpoints.
Pricing: usage is billed in Exa API credits (1 credit = $1.00 USD). Credits are consumed across Search, Contents, Research, and other Exa APIs according to the Exa rate card at https://exa.ai/pricing. Prepaid contracts and private offers commit a credit balance (e.g. a $2,000 contract = 2,000 credits).
For custom pricing, EULA, or a private contract, please contact sales@exa.ai.
Highlights
One API for search, crawling, and research, built for AI agents
Business-grade web search and content retrieval at any scale
Structured, up-to-date web data your applications can rely on
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing is based on the duration and terms of your contract with the vendor. This entitles you to a specified quantity of use for the contract duration. If you choose not to renew or replace your contract before it ends, access to these entitlements will expire.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
This listing offers one pricing dimension: the ExaSearch Contract, billed in units. You commit to a contract for search API access rather than paying per request as you go. Buying through this single dimension gives you a set unit quantity under one agreement. There are no separate tiers, instance sizes, or add-on dimensions to choose between here. To size your commitment or arrange volume terms, work directly with the vendor.
Top-of-mind questions for buyers
What counts as one ExaSearch unit for billing under this contract?
Units map to search API usage. The API meters requests, page contents retrieved, and research runs. Web search requests and deep search requests bill per thousand requests. Full page content retrieval bills per thousand pages. Async research runs bill per run, based on the effort level you set.
What drives my usage under this contract — search requests, page contents, or research runs?
Several activity types draw down your contract. Web search and deep search bill per thousand requests. Page content retrieval bills per thousand pages. Async agent runs bill per run, scaling with compute units, search tool calls, and any contact enrichment. All applicable charges apply together as you use the API.
How does cost change when I run deeper research versus fast searches?
Fast web search and page retrieval bill at set per-thousand rates. Deep search and reasoning workflows cost more per thousand requests because they run multi-step processes. Async agent runs can use a fixed effort mode for predictable per-request pricing, or auto mode where compute scales to the task.
exa.ai+1
Helpful?
Vendor refund policy
All fees are non-cancellable and non-refundable except as required by law.
How can we make this page better?
Tell us how we can improve this page, or report an issue with this product.
Give us feedbackReport a problem with this product or seller
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA).
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
API-Based Agents and Tools integrate through standard web protocols. Your applications can make API calls to access agent capabilities and receive responses.
Additional details
Usage instructions
MCP
API
Exa provides an official Model Context Protocol (MCP) server, allowing MCP-compatible assistants and agents (including Codex, Cursor, Claude Code, and any other MCP client) to use Exa as a native tool.
Configure your MCP client to connect to <https://mcp.exa.ai/mcp>
Authenticate using an Exa API key created in the Exa Dashboard (<https://dashboard.exa.ai>). All MCP requests consume Exa API credits from your organization's prepaid balance according to the standard API pricing.
Email support is available Monday through Friday at support@exa.ai. Documentation, API references, and integration guides are available at https://docs.exa.ai. For custom pricing, enterprise agreements, EULA, or private contracts, contact sales@exa.ai.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
exaBase Studio is a Kubernetes-based, all-in-one platform to design, develop, and run AI agents for business automation - using customizable templates and full-code flexibility.
This service is available for corporate purchase only.
ExaVault is an on-premise file transfer and file sharing appliance for secure and efficient data exchange within and between organizations. It supports multiple protocols, including FTP, SFTP, and WebDAV, and features an intuitive web interface. ExaVault is for businesses seeking a reliable method for managing file transfers while maintaining control over their data.
exaBase Studio for EKS is a low-code/no-code platform that accelerates cloud-native development on AWS EKS. It offers an integrated GUI for microservice design, CI/CD, and auto-scaling configuration, dramatically streamlining the entire process from development through operations.
High-Signal Neural Search with Snappy Performance and Easy Onboarding
Reviewed on Jun 10, 2026
Review provided by G2
What do you like best about the product?
The quality of the search results is easily the standout feature. When you’re feeding data into LLMs, the output is only as good as the retrieval step, and Exa’s neural search consistently surfaces high-signal content that traditional APIs tend to miss. On top of that, performance feels incredibly snappy, and the API documentation made onboarding straightforward. Overall, it’s a major upgrade for any AI-driven search workflow.
What do you dislike about the product?
When you’re building autonomous agents that run in continuous loops, rate limits and API costs can quickly become a bottleneck as you scale. It’s fantastic for prototyping, but once you deploy agents that need constant, high-volume context retrieval, you have to pay close attention to optimization to keep usage costs under control.
What problems is the product solving and how is that benefiting you?
We struggled to feed high-quality, highly relevant context into our autonomous coding agents. When we were building reinforcement learning environments or SWE-agents, traditional search APIs produced too much noise and often returned poorly formatted HTML. Exa addresses this by offering semantic, neural search that delivers clean, structured data directly to our LLMs. The result is noticeably better agent accuracy, along with hours saved because we no longer have to build and maintain custom web scrapers.