E

Exa

by Exa Labs Inc.

AI Search InfrastructureAgent InfrastructureAI Data Infrastructure

Neural, embeddings-based web search API for AI apps

Usage-based · Free tier · Enterprise·Added July 2, 2026·Updated July 2, 2026
Share:
THE DAILY BRIEF
Exa

by Exa Labs Inc.

AI Search InfrastructureAgent InfrastructureAI Data Infrastructure

Neural, embeddings-based web search API for AI apps

Usage-based · Free tier · Enterprise

Exa (formerly Metaphor) is a neural, embeddings-based web search API that gives AI applications and agents high-quality semantic search, page contents, and structured web data.

At a Glance

Category
AI Search Infrastructure
Pricing
Usage-based, Free tier, Enterprise
Target Market
Developers, Startups, Enterprise
Founded
2021
Headquarters
San Francisco, USA

Key Features

  • Neural Search API
  • Contents API
  • Websets
  • Answer / Deep Search
  • Monitors

Capabilities

api access
web search
semantic search
data extraction

Use Cases

  • AI agent web search
  • Company and lead enrichment
  • Cited research assistants

Ideal For

Best For

  • Semantic web search for AI agents
  • RAG pipelines needing high-relevance results
  • Building structured web datasets and enrichment

Not Ideal For

  • Teams that need a full headless-browser scraper for JavaScript form interactions

Pricing

Free

$0

  • Free monthly request allotment
  • Access to Search, Contents, and Answer
  • Configurable latency

Pay-as-you-go

Usage-based

  • Search from $7 per 1,000 requests
  • Contents $1 per 1,000 pages
  • Deep Search $12-$15 per 1,000 requests

Enterprise

Custom

  • Up to 1,000 results per search
  • Custom rate limits
  • Dedicated support

Pay-as-you-go pricing per the official pricing page: Search from $7 per 1,000 requests (up to 10 results), Contents $1 per 1,000 pages, AI summaries $1 per 1,000 pages, Deep Search $12-$15 per 1,000 requests. A free monthly request allotment is included; Enterprise is custom-priced.

THE DAILY BRIEF

Enterprise AI insights for technology and business leaders, twice weekly.

beri.net

Subscribe at beri.net/subscribe for twice-weekly AI insights delivered to your inbox.

LinkedIn: linkedin.com/in/rberi  |  X: x.com/rajeshberi

© 2026 Rajesh Beri. All rights reserved.

Exa (formerly Metaphor) is a neural, embeddings-based web search API that gives AI applications and agents high-quality semantic search, page contents, and structured web data.

At a Glance

Category
AI Search Infrastructure
Pricing
Usage-based, Free tier, Enterprise
Target Market
Developers, Startups, Enterprise
Founded
2021
Headquarters
San Francisco, USA

Key Features

  • Neural Search API

    Embeddings-based semantic search over the web, optimized for AI agents and developers.

  • Contents API

    Retrieve clean, token-efficient page text, highlights, and AI summaries for LLM context.

  • Websets

    Build curated, structured datasets of web entities such as companies, people, and papers.

  • Answer / Deep Search

    Multi-step research workflows that return structured outputs with citations.

  • Monitors

    Track web events at a specified cadence and deliver updates via webhooks.

Capabilities

api access
web search
semantic search
data extraction

Use Cases

  • AI agent web search

    Give agents high-relevance semantic search plus full page contents in one call.

  • Company and lead enrichment

    Use Websets to assemble structured lists of entities and enrich them from the web.

  • Cited research assistants

    Power research tools with Deep Search that return structured, source-cited answers.

Ideal For

Best For

  • Semantic web search for AI agents
  • RAG pipelines needing high-relevance results
  • Building structured web datasets and enrichment

Not Ideal For

  • Teams that need a full headless-browser scraper for JavaScript form interactions

Integrations

API Support
SDK Available
SDK:PythonJavaScript

Deployment

Self-Hosted
Cloud-Hosted
On-Premise
Cloud API (hosted)

Pricing

Free Trial Available

Free

$0

  • Free monthly request allotment
  • Access to Search, Contents, and Answer
  • Configurable latency

Pay-as-you-go

Usage-based

  • Search from $7 per 1,000 requests
  • Contents $1 per 1,000 pages
  • Deep Search $12-$15 per 1,000 requests

Enterprise

Custom

  • Up to 1,000 results per search
  • Custom rate limits
  • Dedicated support

Pay-as-you-go pricing per the official pricing page: Search from $7 per 1,000 requests (up to 10 results), Contents $1 per 1,000 pages, AI summaries $1 per 1,000 pages, Deep Search $12-$15 per 1,000 requests. A free monthly request allotment is included; Enterprise is custom-priced.

Connect

Newsletter

Stay Ahead of the Curve

Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.

Subscribe