Problem you have encountered:
We are experiencing significantly increased latency issues with Gemini API starting from September 14th at 6:00 PM (KST). The response times have been consistently slower than normal, affecting our application’s performance. This issue is occurring with both gemini-2.5-flash models that we are using in our production environment.
What you expected to happen:
We expected the Gemini API to maintain consistent and stable response times as it had been performing previously. The API should provide responses within the normal latency range without any significant delays or performance degradation.
Steps to reproduce:
Make API calls to gemini-2.5-flash models
Monitor the response times and latency metrics
Compare the current latency with the baseline performance before September 14th, 6:00 PM
The increased latency can be consistently observed across multiple API requests
Other information (workarounds you have tried, documentation consulted, etc):
The issue has been persistent since September 14th at 6:00 PM KST
model (gemini-2.5-flash) are equally affected
No changes were made to our implementation or request patterns before the issue started
We have monitoring data/graphs showing the clear increase in latency starting from the mentioned timestamp
This is impacting our production services and user experience
We have not found any relevant service disruption notices or documentation addressing this specific issue