Exa
Officially listedNeural search API platform built for AI systems and LLM applications with zero data retention.
Exa
Exa is a neural search API built for LLMs and AI agents, developed by Exa Labs (formerly Metaphor Systems) and backed by an $85 million Series B from Benchmark, NVIDIA, OpenAI, and other top investors. It is not a wrapper around a search engine but AI-native infrastructure with its own index and embedding models, already adopted by OpenAI, Cursor, and Notion AI.
Core Capabilities
- Neural semantic search: In-house embedding models understand query intent rather than matching keywords, achieving 94.9% semantic retrieval accuracy
- Clean content extraction: Returns parsed plain text/Markdown, minimizing LLM token consumption and lowering hallucination risk
- Ultra-fast responses: The Exa Instant endpoint has P50 latency under 200ms, meeting real-time agent workflow needs
- Deep research mode: The
/researchendpoint automatically runs multi-round searches and outputs structured JSON with citations - FindSimilar discovery: Finds semantically similar pages from a URL — a capability keyword search simply cannot deliver
- Websets data extraction: Describe your need in natural language and automatically generate structured company/people lists with enriched fields
Use Cases Engineers building RAG pipelines and AI agents, product teams injecting real-time web knowledge into LLM applications, enterprise customers in finance and healthcare with strict ZDR privacy-compliance requirements, and sales teams replacing ZoomInfo with Websets for market research.
Unique Advantages The industry consensus is to use Exa for research. Tavily is faster (187ms vs 350ms) and Perplexity is better for trending news, but on semantic search quality (81% vs 71%) and zero data retention, no competitor goes head-to-head with Exa. Websets is a differentiated product in a class of its own.
Editor's Review ARR is up 1010% year over year, and customers include OpenAI and Cursor — that says it all. Users report a 40% improvement in query quality after integrating, and Websets speeds up market research 15-20x. The $10 free credit is enough to thoroughly test core features. The one caveat: at scale, multi-dimensional billing stacks up quickly, so plan your budget ahead.
Pricing
### 定价模式：免费增值（按量付费） **起步价**：免费（注册赠送 $10 额度，无需信用卡） #### API 按量付费方案 - **Search（1-25 结果）**：$5 / 千次 - **Search（26-100 结果）**：$25 / 千次 - **Contents 内容解析**：$1 / 千页 - **Answer 直接问答**：$5 / 千次 - **exa-research 深度研究**：$5 / 千次（+ 页面读取 $5/千页） #### Websets 订阅方案 - **Starter**：$49/月 - 8,000 积分，1 席位，100 结果/Webset - **Pro**：$449/月 - 100,000 积分，10 席位，1,000 结果/Webset - **Enterprise**：定制 - 无限结果/席位，ZDR，SLA，SOC-2 合规 #### 试用/其他信息 新用户注册即获 $10 免费额度，学术机构和初创公司可申请价格优惠。 — Visit website
FAQ
Is Exa free?
Sign-up includes $10 in free credit, no credit card required. Normal usage is pay-as-you-go, with the search API starting at $5 per 1,000 searches.
What can Exa be used for?
Exa offers five APIs — semantic search, web content extraction, similar page discovery, direct question answering, and deep research — with the core purpose of giving LLM applications and AI agents real-time web knowledge retrieval. The Websets product targets market research and sales prospecting.
How does Exa differ from Tavily?
Tavily is faster (187ms), better documented, and more affordable, suiting general RAG scenarios. Exa is stronger in semantic search quality (94.9% vs 93.3% SimpleQA) and zero data retention, fitting deep research and enterprise privacy-compliance scenarios.
Which development frameworks does Exa integrate with?
Official support covers LangChain (ExaSearchRetriever), LlamaIndex, and CrewAI integrations, plus Python and JavaScript/TypeScript SDKs and an MCP protocol server that plugs directly into MCP clients like Claude Desktop.
Is Exa suitable for building RAG applications?
Very much so. Exa returns clean Markdown/HTML content that minimizes token consumption, with 94.9% semantic retrieval accuracy — a top search backend for RAG pipelines. OpenAI has made it the default search tool for its gpt-oss models.
What does Exa's zero data retention (ZDR) mean?
Exa builds its own search index rather than relying on Google/Bing, enabling true zero data retention — queries are not logged or used for training. This is a key selection factor for highly regulated industries like finance, healthcare, and government, and it is SOC-2 certified.
What is Exa Websets, and how does it differ from the API?
Websets is a standalone product for non-technical users: describe your need in natural language and it generates structured company/people data lists with field enrichment (emails, tags, etc.). The API targets developers while Websets targets market research and sales teams — different positioning.