Hi everyone,
Over the past few days, we have been experiencing several issues while working with the Gemini API in production:
-
Frequent
429 Resource Exhaustedand rate-limit errors even under moderate traffic -
Sudden response delays and timeout issues
-
Inconsistent outputs from the same prompts
-
Occasional API failures during function/tool calling
-
Higher latency after recent model updates
We also noticed that some developers are discussing similar instability and API-related problems across the community.
Is Google currently facing any known infrastructure or scaling issues with Gemini API? Are there any official recommendations for handling these recent API performance problems in production environments?
Would appreciate any updates or workarounds from the team or other developers.
Thank you!