Gemini 3.1 Flash Live Preview returns quota exceeded on every new API key

Hi Google AI Team,

I’m facing a persistent quota issue with `gemini-3.1-flash-live-preview`.

The model appears in `client.models.list()`, and I can successfully connect using the Live API with `response_modalities=[“AUDIO”]`. The connection prints `CONNECTED`, so the Live API endpoint is reachable.

However, when the model starts generating a response, I receive:

```

google.genai.errors.APIError: 1011 None. You exceeded your current quota, please check your plan and billing details.

livekit.agents._exceptions.APIConnectionError: Failed to connect to Gemini Live after 3 attempts

```

I tested multiple newly created API keys from different Google accounts, and the same quota exceeded error occurs almost immediately, even though the keys are new and have barely been used.

I also confirmed that:

* `gemini-2.5-flash-native-audio-latest` connects successfully.

* `gemini-3.1-flash-live-preview` appears in the available models.

* The issue happens during actual response generation.

Is this a quota allocation issue specific to `gemini-3.1-flash-live-preview`, an account-level restriction, or a regional limitation?

Any guidance would be greatly appreciated. Thank you.