Skip to main content
All V4 responses are cached in-memory upstream with endpoint-specific TTLs. Concurrent requests for the same upstream URL share a single fetch (request coalescing), so high concurrency does not cause upstream amplification.

TTL Reference

EndpointTTLNotes
/api/v1/sports1hRarely changes
/api/v1/bookmakers1h
/api/v1/bookmakers/top1h
/api/v1/bookmakers/payouts1h
/api/v1/navigation1h
/api/v1/matches (inplay)30sFast refresh for live
/api/v1/matches (other)2m
/api/v1/matches/smart-bets2m
/api/v1/odds/max (inplay)30s
/api/v1/odds/max (other)2m
/api/v1/polls5m
/api/v1/tipsters10m
/api/v1/match/{id}1m
/api/v1/match/{id}/odds1m
/api/v1/match/{id}/odds-movements2m
/api/v1/match/{id}/markets5m
/api/v1/match/{id}/prediction5m
/api/v1/leagues10m
/api/v1/categories10m
/api/v1/league/{id}/standings10m
/api/v1/league/{id}/top-scorers10m
/api/v1/league/{id}/outrights5m
/api/v1/league/{id}/special-offers1h
/api/v1/team/{id}10m
/api/v1/team/{id}/matches5m
/api/v1/team/{id}/offers1h
/api/v1/news10m
/api/v1/world-cup2m / 10mMatches 2m, standings/outrights 10m
/api/v1/world-cup/matches2m
/api/v1/world-cup/standings10m
/api/v1/world-cup/outrights5m
/api/v1/world-cup/predictions2m / 10m
/api/v1/world-cup/special-offers1h
/api/v1/world-cup/news10m

Stale Fallback

If the upstream is briefly unavailable, the last-known cached payload continues to be served so your UI never goes blank. Check the generatedAt field if you need to detect staleness in your application.
Last modified on June 11, 2026