Skip to content
AI Primer
breaking

Google says Gemini API spend caps now enforce within 10 minutes

Google says Gemini API spending caps are enforced with a ten-minute latency, down from more than a day. Real-time enforcement remains in development.

2 min read
Google says Gemini API spend caps now enforce within 10 minutes
Google says Gemini API spend caps now enforce within 10 minutes

TL;DR

Google already described the same ten-minute delay in an April forum reply. Long-running batch jobs and agent sessions can also exceed a project's cap.

Ten-minute enforcement

Google says the current implementation carries an experimental label and discloses the delay in the interface. Ten minutes is a hefty asterisk on a spending cap.

The stated improvement from more than a day to ten minutes comes without a rollout date in the reply. Google says the team is working on a real-time version.

The apology drew a refund demand from AIandDesign, who asked: “If we cannot trust these caps then what are we doing?”

Budget alerts versus spend caps

Google distinguished two controls in the exchange:

  • Budget alerts: emails notifying users about spending.
  • Gemini API spend caps: enforced limits, currently subject to approximately ten minutes of latency.

The project-cap documentation explicitly says billing data can take around ten minutes to process, allowing additional charges to accrue beyond the configured amount.

Monthly caps in AI Studio

Google's billing guide describes how the controls are configured and shared:

  • Location: AI Studio's Spend page, under Monthly spend cap → Edit spend cap.
  • Permissions: project editors, owners, and admins can set project caps.
  • API-key scope: every key within a project contributes to the same project cap. Keys have no independent billing settings.
  • Account scope: billing-account tier caps aggregate Gemini API spending across linked projects. Reaching that limit pauses service for all those projects until the next monthly billing cycle.
  • Exclusions: spend caps are unavailable for invoiced or offline accounts.

The guide separately warns that total-cost breakdown charts on the Billing and Spend pages can take up to 24 hours to update. That display delay is distinct from the roughly ten-minute billing-processing delay behind cap enforcement.

Further reading

Discussion across the web

Where this story is being discussed, in original context.

On X· 2 threads
TL;DR1 post
Ten-minute enforcement1 post
Share on X