Exa-Search

Exa-Search

Provider: Exa
Category: Research & Search
Endpoint: POST /v1/search
Context window:
Served from:

Web search engine for finding pages, retrieving similar pages, crawling, and dedicated code search across the open web for AI agents.

At a glance

FieldValue
Model idexa-search
Input modalitiestext
Output modalitiestext
Context window
Region
Featuresweb_search, semantic_search
NewNo
Native inferenceNo

Pricing

ChargeSpecRate
Search (1-25 results)per search$0.0060
Search (26-100 results)per search$0.030
Content (Text/Highlights/Summary)per page/feature$0.0060
Code Searchper 1k tokens$0.0060

Example request

$curl https://api.empiriolabs.ai/v1/search \
> -H 'Authorization: Bearer $EMPIRIOLABS_API_KEY' \
> -H 'Content-Type: application/json' \
> -d '{"model": "exa-search", "query": "latest LLM benchmarks 2026"}'

Parameters

ParameterTypeRequiredDefaultDescription
querystringyesSearch query, or URL for ‘similar’ operation, or repo path for ‘code’ operation.
operationenumno"search"search: keyword/neural search. similar: pages similar to the URL in query. contents: fetch full text of given URLs. code: code-aware search. · Allowed: search, similar, contents, code
search_typeenumno"auto"auto: Exa picks. instant: cheapest, fastest. neural: semantic search. deep: thorough. fast: speed-optimized. · Allowed: auto, instant, neural, deep, fast
num_resultsnumberno10Number of results to return. · Range: 1 – 100
categoryenumnoRestrict by source category. · Allowed: “, company, research paper, news, pdf, github, tweet, personal site, linkedin profile, financial report
include_domainsstringnoComma-separated allowlist of domains.
exclude_domainsstringnoComma-separated denylist of domains.
include_textstringno5-word filter — only return pages that contain this exact phrase.
exclude_textstringno5-word filter — drop pages that contain this exact phrase.
start_published_datestringnoISO 8601 date — only pages published after this.
end_published_datestringnoISO 8601 date — only pages published before this.
start_crawl_datestringnoISO 8601 — only pages crawled after this.
end_crawl_datestringnoISO 8601 — only pages crawled before this.
show_contentbooleannotrueShow page content snippets in results.
textbooleannofalseReturn full page text. Honored when show_content=true.
text_max_charsnumbernoChar cap on text. Empty = unlimited.
include_html_tagsbooleannofalsePreserve HTML structure in returned text.
highlightsbooleannofalseReturn short highlight snippets.
highlights_sentencesnumberno3Sentences per highlight. · Range: 1 – 10
highlights_per_urlnumberno3Highlights per result. · Range: 1 – 10
highlights_querystringnoCustom phrase to highlight (different from main query).
summarybooleannofalseReturn a per-page summary.
summary_querystringnoCustom prompt for the summary (only when summary=true).
subpagesnumberno0Crawl this many subpages per result. · Range: 0 – 10
subpage_targetstringnoKeyword to bias subpage selection (e.g. ‘pricing’).
livecrawlenumno"fallback"Live-crawl strategy. always: re-fetch every page. fallback: re-fetch only when index is stale. never: index only. preferred: try live then index. · Allowed: always, fallback, never, preferred
code_tokensenumno"dynamic"Code-search only. Token budget per code result. · Allowed: dynamic, 5000, 10000, 20000
disable_formattingbooleannofalseReturn raw upstream response.

Live machine-readable schema is also available at GET https://api.empiriolabs.ai/v1/models/exa-search.