While testing the Realtime API function calls, I sent a function_call_output message with conversation.item.create after each function call. Then, I triggered the server response by sending a response.create message.
After a few function calls, the server returned an error:
{
"type": "error",
"event_id": "event_xxxxx",
"error": {
"type": "invalid_request_error",
"code": null,
"message": "Conversation already has an active response",
"param": null,
"event_id": null
}
}
However, the logs show that all previous responses had received response.done messages from the server, indicating that there were no incomplete responses.