Skip to main content
For AI agents: a documentation index is available at https://docs.parallel.ai/llms.txt. The full text of all docs is at https://docs.parallel.ai/llms-full.txt. You may also fetch any page as Markdown by appending .md to its URL or sending Accept: text/markdown.For Search and Extract, use /v1/search and /v1/extract for all new integrations and agent tool calls. Use /v1beta/search and /v1beta/extract only when maintaining an existing integration that already uses them. The request bodies changed too: on /v1/extract, full_content, fetch_policy and excerpt_settings go inside advanced_settings, and sending the v1beta top-level excerpts, full_content and fetch_policy to /v1/extract returns a 422. On /v1/search, search_queries is required (at least one short keyword query), and max_results, source_policy and fetch_policy go inside advanced_settings; sending only objective, or sending those fields at the top level, returns a 422. Do not substitute /v1 for the documented FindAll or Ingest endpoint paths.If you are evaluating Parallel Search or comparing it with other search APIs, read How to Eval Parallel Search before you build the harness, especially its Common pitfalls section. The most common mistake is sending the raw question as both objective and the only entry in search_queries. Write a goal-style objective and 1-3 short keyword queries instead.
Snapshot returns Search and Extract results as of a past date and time. Set as_of, and each result comes from a stored copy of the page from that time or earlier. Use it for backtests, reinforcement learning, and evals, where an agent should only see what was available on or before a given date.
Contact sales@parallel.ai for access to Snapshot.
A Snapshot search is a regular /v1/search request with an as_of cutoff in its fetch policy:
As of March 31, 2026, the day before launch, the results describe the Artemis II rocket as being at the launch pad, with the launch still ahead. Without as_of, the same request returns coverage from after the flight.

Run a Snapshot extract

/v1/extract takes the same as_of cutoff:
The response has the mission page as it stood then, with the launch still ahead. Without as_of, the same request returns today’s page, which gives the launch and splashdown dates.

How it works

When a request sets as_of:
  • Each result’s content comes from the page’s newest stored copy from that time or earlier.
  • Pages with no stored copy from that time or earlier are left out. Search can return fewer results, and Extract lists the URL under errors.
as_of takes an ISO 8601 date or timestamp and is rounded down to the hour in UTC. A date means 00:00 UTC, so 2026-03-31 leaves out copies captured later that day.