{"error":{"code":429,"message":"Resource exhausted. Please try again later. Please refer to https://cloud.google.com/vertex-ai/generative-ai/docs/error-code-429 for more details.","status":"RESOURCE_EXHAUSTED"}}

Action Step: Check your API usage and provider status to diagnose the 429 RESOURCE_EXHAUSTED error. - Go to your Vapi dashboard and navigate to Observe > API Logs to review recent requests and spot any quota or rate limit issues. - Also, visit the [Vapi Status Page](https://status.vapi.ai/) and your provider's status page to confirm there are no ongoing outages. This helps pinpoint if the error is due to hitting usage limits or a temporary provider issue. For more details, see the [debugging guide](https://docs.vapi.ai/debugging). Source: - [Debugging voice agents](https://docs.vapi.ai/debugging)

I've been testing knowledge bases using uploaded files and about half of the time the tool calls return this error. I would try other providers but it looks like google is the only one allowed. I have been using the web chat and talk to assistant features when I see this, is there a difference between this and prod? We cant use knowledge bases in production if its this unreliable during testing.

Hi Pat, could you please provide me the call IDs?

Sure! Heres an example 7146ee20-3ffc-4c1f-abb4-1737e5b80b03

Hi Pat, apologies for the delay. Vapi's shared Google Vertex AI credentials are hitting rate limits, which is causing the 429 errors you're seeing on knowledge base queries. This is a platform-level issue affecting multiple orgs and not specific to your setup. We're actively looking into a fix. In the meantime, please let us know if you continue to see these errors so we can track the frequency. Sorry for the inconvenience! Best, Vapi Support

I am experiencing this issue today. Has it not been fixed?

Hi, could you please share the call IDs?