Exa is a modern AI search engine with SERP API, website crawler tools, and deep research API. Power your app with web search AI and web crawling API.
/agent

A single API for frontier web researchEnrich entities, build large lists, and run deep research from a single API with built in structured results and web citation

Return structured intelligence on this ad partner: brand partnerships, customer stories, and NVIDIA/cloud investments.
Exa Agent API resultAttribute Enrichment
Company
Brand Partnerships
Customer Stories
NVIDIA / Cloud
1
Amazon.com
AWS-NVIDIA full-stack collab
100K+ customers on Bedrock
Blackwell HGX B300/GB300
2
Meta Platforms
Multiyear NVIDIA partnership
GEM: 3.5× FB ad clicks
Millions of Blackwell GPUs
3
Netflix
Dove × Bridgerton, Amazon DSP
>1B impressions, 60% new shoppers
Netflix Ads Suite in-house
4
Alphabet
AI Hypercomputer, Vera Rubin
Home Depot: 2-digit ad lift
First Vera Rubin NVL72
5
Snowflake
$200M OpenAI, $6B AWS
Canva: visual AI at scale
NVIDIA NeMo in Cortex

Built for real-world use cases

CRM Enrichment, GTM Intelligence, KYC/KYB insights, research reports and more across startups and fortune 500 teams

Use cases
Try in API Playground
Query

Company
Brand Partnerships
Customer Stories
NVIDIA / Cloud Investments
1
Amazon.com
2
Meta Platforms
3
Netflix
4
Alphabet
5
Snowflake
6
Cisco Systems

Frontier quality at a fraction of the cost

Exa Agent outperforms comparable systems across web research, entity enrichment, and financial analysis benchmarks.Our goal was to produce the best web research agent using methodologies like model fusion and Exa's token-efficient highlights model (which have shown up to 94% reductions in token usage). The result is significantly lower cost and latency for frontier performance.

Low36.2 / $0.025
Medium43 / $0.1
High52 / $0.5
Perplexity Pro26.5 / $0.1
Perplexity Advanced49.5 / $0.9
Parallel Ultra39.4 / $0.3
Opus 4.852.4 / $2.16
GPT 5.551.8 / $1.9
Score / Cost

Pay per request, scale to any effort level

$0.01–$1+/request
minimal$0.012

Lightweight tasks, lowest cost

low$0.025

Simple lookups, narrow factual tasks, short answers

medium$0.10

Default starting point for most standard research tasks

high$0.50

Harder research, more citations, stricter completeness

xhigh$1.00

High-value tasks where completeness matters more than cost or latency

autodynamic

Dynamically scales compute to match the complexity of the task