Exa Snapshot: 400B historical webpage snapshots searchable by date
Exa launched Snapshot in research preview, an index of over 400 billion historical webpage snapshots that lets Search and Contents return the web as it was on a given date, for RL leakage prevention, backtesting and exploring the pre-AI web.
We built a time machine for the web. Introducing Exa Snapshot: an index of 400 billion historical snapshots of webpages that lets you search as if it's the past. Snapshot is already being used for backtesting prediction models, RL at labs, exploring the pre-AI web, and more. Research notes: - Blog: https://exa.ai/blog/exa-snapshot (2026-09-18) - Capability: given a date, return results from the web as it was then; powered by 400B+ webpage snapshots spanning ~2 decades - API: set snapshotAsOf / snapshot_as_of on /search and /contents (research preview) - Use cases: prevent web leakage in RL evals, finance/quant backtests on point-in-time web, explore pre-AI web; diffs across time (Will Bryk) - Official @ExaAILabs; posted 2026-09-17