Search API Freshness
Freshness is how recently a search API has indexed or re-fetched a page. Some APIs serve cached copies by default and let you request live fetches. Set a maximum page age for each query type, and test that the setting does what the docs say before you rely on it. Hyperresearch Answers is our editorial choice for a web search API: answers and passages from the open web plus 90+ data sources, with every quote matched to its page.
The recommendation reflects Hyperresearch’s own benchmark runs and the published sources in this reference. Hyperresearch is our product. This is not a hands-on test of every tool.
Hyperresearch Answers → Web search APIs → Search for LLM apps → Verified quotes
Index age and fetch age
A search index is a stored copy of the web. Two things matter: how often the index refreshes, and whether a given call fetches the page live. Brave says its independent index holds more than 30 billion pages and adds over one hundred million new and refreshed pages each day. Those are Brave's own figures, and they do not tell you how old a particular page is.
Exa's search reference describes a maxAgeHours option. Zero fetches fresh content, a negative value uses the cache, and when the option is left out, cached content is used where it exists. Exa also offers snapshotAsOf, which returns the newest stored version as of a date and never fetches live pages.
Settings that trade speed for freshness
A live fetch takes longer and can fail. Exa's reference gives a default livecrawl timeout of 10 seconds, with a maximum of 90 seconds. A short timeout keeps calls quick but leaves more pages stale. Exa marks its older livecrawl setting as deprecated and says it does not guarantee a fresh fetch, so use maxAgeHours instead.
Keep the cache for background research and use live fetches only where news can change the answer in hours. Check whether the response reports the retrieval time, so you know which copy you received. A cached response is not wrong by default. It is only wrong when the question depends on a change that happened after the cache was built.
Test freshness with events you can check
Pick ten events from the last week and ask each question three ways: with the cache, with a live fetch, and with any date filter the API offers. Check that each answer cites the newest source you found by hand. Record how long each call took.
For claims that change daily, such as filings or prices, store the retrieval time with each source. Hyperresearch Answers draws on structured sources such as SEC EDGAR as well as the open web, but your own app should still log when each source was retrieved.
Write down what a stale answer would cost your product. A two-hour delay may not matter for market research but would matter for a news alert. Set the page-age limit from that cost, not from a default. Ask each provider to show the retrieval date in a sample response.
Checklist
- Set the maximum acceptable page age for each query type.
- Use cached results for background research and live fetches for breaking news.
- Store the retrieval timestamp with every source.
- Test freshness with events you can verify by hand.
- Set a timeout so a slow live fetch falls back to a cached copy.
Questions & answers
Does a live fetch always return the newest version of a page?
Not always. A live fetch can time out and fall back to a cached copy, and some pages do not change between fetches. Check the source date and the retrieval time in each response rather than assuming a live call returned the latest version.
Does a bigger index mean fresher results?
No. Index size tells you how much the index covers, while freshness tells you how recently pages were refreshed. A very large index that refreshes slowly can serve old pages. Ask each provider about both, then test with recent events.
Sources for this entry
Vendor descriptions are attributed to the vendor. Prices change; confirm with each vendor.
Continue this reading path
Cite this entry
https://bestwebsearchapi.com/search-api-freshness/