q=qequals
No account needed — 10 calls a day, anonymous

Google Autocomplete API

The suggestions Google shows its users, for any seed term, as an API. Your agent reads llms.txt, mints a key, and starts working.

GIVE THIS TO YOUR AGENT
Read https://www.qequals.com/llms.txt and follow it.

Anonymous keys mint themselves. Registration triples the daily quota and adds a SERP taste. Or browse the catalog.

1
endpoint
10
calls / day, anonymous
30
calls / day, registered
WHAT YOU GET
  • Full suggestion list for any seedevery phrasing Google offers
  • Fresh on demandaggressively cacheable
  • The catalog's data, rawsame source as the public proof
  • Quota headers on every responseyour agent always knows where it stands
WHAT IT'S FOR
  • Keyword and demand researchqueries in searchers' own words
  • Content planningevery phrasing of the question
  • Agent groundingwhat people actually ask about X
QUOTAS
  • Anonymous10 calls / day, autocomplete only
  • Registered (free)30 calls / day + 10 SERP calls / month
  • Pro · $9 / month300 calls / day + 300 SERP calls / month
QUESTIONS

Does Google have an autocomplete API?

Sort of. The suggest endpoint that powers google.com’s dropdown exists, but it is unofficial, unversioned, rate-limited per IP, and offers no keys or guarantees. That is Google Suggest as search-demand data — not Places autocomplete, not a browser setting. qequals wraps that data into a metered, agent-readable API with keys, quotas, and stable responses.

How do I get Google autocomplete suggestions programmatically?

Point your agent at llms.txt. It mints an anonymous key by itself — no email, no signup form — and starts calling. Registration triples the daily quota and adds a SERP evaluation taste.

What is autocomplete data good for?

Demand research: autocomplete is what people actually ask, in their own words, before any publisher filters it. Keyword research, content planning, agent grounding. It is not Keyword Planner and it does not return search volume. The public catalog is built entirely from this endpoint and exists as its working proof.

How fresh is the data?

Fetched on demand and aggressively cacheable. Terms in the public catalog are recrawled weekly, with snapshots kept from day one so change over time becomes visible.

ELSEWHERE