ExaWebAvailable now

exa/contents

exa/contents is a web tool from Exa that runs in the background and returns structured JSON. Pay as you go, with one API key.

Run time p50 · last 24 hours
3.4s
Success rate · last 24 hours
100.0%

Pricing

$0.0013 per page for each content type (text, highlights, summary), subpages included.

Limits

Runs
In the background
Returns
json
Longest queue wait
2 minutes
Run deadline
5 minutes
Longest run, end to end
167 minutes
Largest inline input
1 MB
Largest input file
500 MB
Largest output
5 GB
Outputs per run
16

Input

FieldKindDefaultDescription
idslist—

Document IDs obtained from searches.

textJSON—

Text extraction options for each result.

urlslist—

URLs to crawl (backwards compatible with the ids parameter).

extrasJSON—

Extra parameters to pass.

contextJSON—

Deprecated. Use highlights or text instead. Returns page contents as a combined context string.

summaryJSON—

Summary of the webpage.

subpagesnumber (whole)
0 to 100
0

The number of subpages to crawl. The actual number crawled may be limited by system constraints.

livecrawlone of
never, always, fallback, preferred
—

Deprecated. Use maxAgeHours instead. Does not guarantee freshly fetched content. Cannot be combined with maxAgeHours.

complianceone of
hipaa
—

Enterprise-only compliance mode. Set to hipaa for HIPAA mode. Requires cache-only retrieval with supported parameters. See the HIPAA docs for details.

highlightsJSON—

Text snippets the LLM identifies as most relevant from each page.

maxAgeHoursnumber (whole)
-1 to 720
—

Maximum age of cached content, in hours, up to 720. Cached content younger than this is used, and older pages are fetched fresh. 0 always fetches fresh content and -1 always uses the cache. When omitted, cached content is used when available and pages without it are fetched live.

snapshotAsOfJSON—

Return the newest stored version of each page as of this instant. Accepts an ISO 8601 date-time (2026-01-15T00:00:00Z) or a date-only value (2026-01-15), which is interpreted as midnight UTC. Historical requests are served from stored versions only and never fetch the live page.

subpageTargetJSON—

Term to find specific subpages of search results. Can be a single string or an array of strings.

livecrawlTimeoutnumber (whole)
… to 90000
10000

The timeout for livecrawling in milliseconds.

crawledBeforeDatetext—

Deprecated. Use snapshotAsOf instead. Alias of snapshotAsOf with identical behavior.

Same request in code

curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "exa",
    "model": "exa/contents",
    "input": {}
  }'