StartupSan Francisco, CA · Est. 2021
Exa
Search API purpose-built for AI agents.
Exa (formerly Metaphor) provides neural search and web-crawl APIs tuned for agent consumption — live, cited knowledge for agents instead of stale training data.
Pros & cons
Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.
Pros
- User reportsRuns its own large, independent web index (confirmed on HN by Exa's head of index) rather than reselling Google or Bing.[1]
- User reportsIn one user's side-by-side test, LLMs preferred Exa and Tavily results over Brave and SearXNG.[1]
- User reportsBuilt into agent tools: OpenCode supports Exa search by default, and Exa is available as a Codex plugin.[1][2]
- DocsThe free Starter plan gives $10 of credits a month plus a $10 onboarding bonus, with no payment method needed.[1]
Cons
- DocsPricier than some rivals: search starts at $4 per 1,000 requests (users quoted $7 per 1,000 plus a charge per result beyond the top 10), against $0.25-5 elsewhere.[1][2][3]
- DocsContents cost $1 per 1,000 pages for each content type, so asking for text, highlights and summaries multiplies the cost.[1]
- DocsZero data retention and HIPAA are Enterprise-only, which one user described as Exa charging for no retention.[1][2]
- DocsThroughput is capped at 10 search QPS on the free plan and 25 on Developer; higher QPS and SLAs are Enterprise-only.[1]
- Best for
- Agents that need high-quality semantic web search from an independent index.
- Watch out for
- Model your costs per request and per content type, and get zero retention in writing if your queries are sensitive.
Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction
People
Open roles (67)
- Product Manager, Agentic Search
San Francisco, California · Full-time · Posted Nov 20, 2026
Product & design - Product Partnerships
San Francisco / New York · Full-time · Posted Nov 19, 2026
Go-to-market - Strategic Finance
San Francisco, California · Full-time · Posted Nov 16, 2026
Operations - Head of Product Marketing
San Francisco, California · Full-time · Posted Nov 13, 2026
Go-to-market - Safety Engineer - Abuse and Fraud
San Francisco, California · Full-time · Posted Nov 12, 2026
Engineering
On X
No posts from them in our X feed yet.
Videos

Cognition · Nov 19, 2026
How Exa Uses Devin Bots
Cost & limits
Neighbours in the directory
San Francisco, CA · Est. 2022
The data framework for connecting agents to your data.
LlamaIndex provides ingestion, indexing and retrieval primitives that give agents grounded context over enterprise data — a foundational piece of the retrieval side of the harness stack.
San Francisco, CA · Est. 2023
The integration layer: hundreds of managed tools your agents can call.
Composio handles auth, schemas, observability and reliability for the external SaaS tools an agent uses — 'plumbing for agentic actions' across Slack, GitHub, Salesforce and 250+ other surfaces.
Est. 2024
A registry and toolchain for MCP servers.
Smithery indexes and helps deploy MCP (Model Context Protocol) servers — a distribution layer for the tool ecosystem agents connect to.
Connections
Relations from the Harness Hub knowledge graph, as charted in the State of Harness reports.
Similar entries · 1
- TavilyStartup