Hi team, seems like 'model-output' webhook message is not working for GPT realtime model, it doesn't output the model response (except tool calls for some reason), please help to check on this

Suggestion: Based on the documentation, the

model-output

webhook should send tokens or tool-call outputs as the model generates them. If you're not receiving model responses for GPT realtime models, try verifying your webhook configuration and ensure your server is set up to handle the following payload:

json
{
  "message": {
    "type": "model-output",
    "output": { /* token or tool call */ }
  }
}

If the issue persists, this may be a limitation or bug—consider reaching out to Vapi support or checking the [Vapi status page](https://status.vapi.ai/) for updates. Reference: [Vapi server events documentation](https://docs.vapi.ai/server-url/events) Source: - [Vapi server events documentation](https://docs.vapi.ai/server-url/events)

it’s not working only for realtime models, other models are working fine

Our team is looking into it. For now, we recommend using non-realtime models if you are using tools

ah ok thanks, please let me know when the realtime is working

Sounds good.