Vapi is the Voice AI platform for developers.

Vapi AI

Hey I'm seeing this, when using Gemini 2.5 Flash. Any idea what's the issue?

**This error means Google's Gemini LLM service failed during your call—often due to a provider-side issue like a temporary outage or internal error.**  
To proceed, check if the issue persists after retrying the call. If it continues, gather your `call_id` and contact Vapi support for further investigation, as recommended in the [call ended reason documentation](https://docs.vapi.ai/calls/call-ended-reason).

*Example support message:*
```
Subject: Providerfault Google LLM Failed - Gemini 2.5 Flash
Call ID: [your_call_id_here]
Description: Seeing "call.in-progress.error-providerfault-google-llm-failed" when using Gemini 2.5 Flash.
```

Source:
- [Call ended reason documentation](https://docs.vapi.ai/calls/call-ended-reason)

Call id: 019c335a-4bbb-7cca-adfb-4ba122b255b7

HI, thank you for providing the Call ID, that was helpful I reviewed the logs and found errors with the Gemini's API. Could you try using another LLM? it might help fix your issue.

Hi, If you require to use Gemini flash. Could you disable the "Assistant speaks first with model generated message" option and keep "Assistant speaks first"? This error is coming from keeping the model generated message option. We are looking into this issue. But as a workaround, please try to use the "Assistant speaks first" option.