A quiet status page is not a promise that everything works. Labs open an incident once they've confirmed one, and that can be a while after your requests start failing. If you're seeing errors and this page says nothing, you're probably early rather than wrong.
What we can say is what Google published and when, for the Gemini API and Vertex AI.
Google doesn't run a status page just for Gemini. This reads Google Cloud's incident history and keeps the entries Google tagged with a Gemini or Vertex product, so a storage outage in one region doesn't show up here as a Gemini problem. The trade is that a Gemini fault Google files under something else won't reach us either.
# Incident Report
## Summary
On Friday, 27 February 2026 at 04:37 US/Pacific, customers using Vertex AI Gemini API models experienced increased error rates. Impacted services included Google Cloud Support, Agent Assist, Vertex Gemini API and Dialogflow CX in US regions and the global endpoint. The issue persisted for a duration of 1 hour and 58 minutes.
This is not the level of quality and reliability we strive to offer you, and we have taken immediate steps to improve the platform’s performance and availability.
## Root Cause
This incident was caused by a configuration change to a safety…
Affected products: Agent Assist, Dialogflow CX, Vertex Gemini API, Google Cloud Support
If the status page is clear and it's still broken
Check what shipped. A client library or a coding tool that updated this morning explains a surprising number of sudden breakages, and the changelogs list those by the hour.
Then check whether the model you're calling is on its way out. The deprecation list has the retirement dates each lab has published.