Hello, requesting allowlist access for the experimental Lyria RealTime model
on my Gemini API project.
- Project ID: gen-lang-client-04604507..
- Project owner: Juan Bonora,
- Model requested: models/lyria-realtime-exp
- API version: v1alpha
- Preferred region: us-central1
- Billing: enabled (USD 25 prepaid credit on file)
Use case: Building a local prompt-DJ web app on top of Lyria RealTime.
Single-user, single-session, no public deployment. The app exposes
weighted-prompt mixing, BPM/scale/density/brightness controls, and records
sessions to WAV. It's for personal music exploration and testing the model's
steerability characteristics.
Connection symptom today: WebSocket handshake succeeds and `setupComplete`
arrives, but no audio chunks are produced after `play()`. No filtered_prompt
or error frame is sent. This pattern is consistent with the model not being
allowlisted on my project despite billing being active (Lyria 3 generates
fine on the same key).
Thanks.
Hello @J_Bonora,
This forum is primarily intended for the Gemini API and Google AI Studio. For issues related to the Vertex AI API; Contact your sales representative if you have one. Otherwise, you may find more relevant assistance on the Google Cloud Forum