Hi Google team,
I’m building a production AI voice ordering system for restaurants using Twilio Media Streams + Gemini Live API, but I’m hitting a blocker.
Issue:
Getting error 1007 “Request contains an invalid argument” when connecting to the Live API WebSocket endpoint.
Setup:
- Model:
gemini-2.0-flash-exp - Endpoint:
wss://generativelanguage.googleapis.com/ws/google.ai.generativelanguage.v1alpha.GenerativeService.BidiGenerateContent - Billing:
Enabled with $300 credits - Generative Language API:
Enabled - Current quota: Only 2 RPM (too low for production)
Error logs:
The connection establishes successfully, but immediately closes with error 1007 after sending the setup message.
Questions:
- Does my account need beta access to use the Live API?
- Can you increase my quota for
gemini-2.0-flash-expto at least 60 RPM? - Is this model ready for production, or should I wait for a stable release?
Why Gemini:
This is a cost-effective solution for small restaurants. Gemini’s pricing (~$0.02/min) is essential for our business model, as alternatives like OpenAI ($0.30/min) are too expensive.
Any help would be greatly appreciated! Thank you!