I can chat normally with Realtime API but when I need it to return function calling or lookup function calling I get all failed status.
I implement a server to handle twilio event with node.js. But realtime API tends to respond to normal questions, but when function calling is needed, it does not return function calling but the event response.done - status failed.
{
"type": "response.done",
"event_id": "event_C6vrpY4U9amr58QAjvxkW",
"response": {
"object": "realtime.response",
"id": "resp_C6vrlAGoKjcVZDZE3ZNCH",
"status": "failed",
"status_details": {
"type": "failed",
"error": {
"type": "server_error",
"code": null,
"message": "The server had an error while processing your request. Sorry about that! Please contact us through our help center at help.openai.com if the error persists. (include session ID in your message: sess_C6vriUceoMIP9I9DwXtWM). We recommend you retry your request."
}
},
"output": [],
"conversation_id": "conv_C6vripZtIwD0RnTecXh1X",
"modalities": ["audio", "text"],
"voice": "coral",
"output_audio_format": "pcm16",
"temperature": 0.8,
"max_output_tokens": 4096,
"usage": {
"total_tokens": 4638,
"input_tokens": 4351,
"output_tokens": 287,
"input_token_details": {
"text_tokens": 4351,
"audio_tokens": 0,
"image_tokens": 0,
"cached_tokens": 4288,
"cached_tokens_details": {
"text_tokens": 4288,
"audio_tokens": 0,
"image_tokens": 0
}
},
"output_token_details": {
"text_tokens": 287,
"audio_tokens": 0
}
},
"metadata": null
}
}
2 Likes
I’m seeing the same error when trying to call functions with the Realtime API. It was working fine yesterday afternoon (around 1pm PST), but it stopped working last night. No code changes were made in between. Here’s the response I’m getting, similar to OP’s error.
{
"type": "response.done",
"event_id": "event_C72oN8dErqmlbFg03OfbV",
"response": {
"object": "realtime.response",
"id": "resp_C72oMsy2ur2waraypehfu",
"status": "failed",
"status_details": {
"type": "failed",
"error": {
"type": "server_error",
"code": null,
"message": "The server had an error while processing your request. Sorry about that! Please contact us through our help center at help.openai.com if the error persists. (include session ID in your message: sess_C72oJDwpDWIKZZdyl2jlQ). We recommend you retry your request."
}
},
"output": [
{
"id": "item_C72oMGZtM9iCRwdHPIq18",
"object": "realtime.item",
"type": "message",
"status": "completed",
"role": "assistant",
"content": [
{
"type": "audio",
"transcript": "Capturing screenshot. Please give me a moment."
}
]
}
],
"conversation_id": "conv_C72oJn9RgzP91cuBTlD1t",
"modalities": [
"audio",
"text"
],
"voice": "ballad",
"output_audio_format": "pcm16",
"temperature": 0.6,
"max_output_tokens": "inf",
"usage": {
"total_tokens": 509,
"input_tokens": 407,
"output_tokens": 102,
"input_token_details": {
"text_tokens": 389,
"audio_tokens": 18,
"image_tokens": 0,
"cached_tokens": 0,
"cached_tokens_details": {
"text_tokens": 0,
"audio_tokens": 0,
"image_tokens": 0
}
},
"output_token_details": {
"text_tokens": 34,
"audio_tokens": 68
}
},
"metadata": null
}
}
Hi, I’m getting a similar error and it suddenly started happened about the same time as everyone else. I tried changing a couple of session settings and it narrowed down to it erroring every time it tries to function calling. Weirdly, it worked properly (kinda) on the Playground and showed as operational on the status site.
I reached about to the support already though but they haven’t replied yet.
{
"object": "realtime.response",
"id": "resp_C7Hq8DyHL0xjJI5oaheJz",
"status": "failed",
"status_details": {
"type": "failed",
"error": {
"type": "server_error",
"code": null,
"message": "The server had an error while processing your request. Sorry about that! Please contact us through our help center at help.openai.com if the error persists. (include session ID in your message: sess_C7HpvYwxJT5CgpmFyxrSA). We recommend you retry your request."
}
},
"output": [],
"conversation_id": "conv_C7HpwN9Nax8pzInBZgf82",
"modalities": [
"audio",
"text"
],
"voice": "sage",
"output_audio_format": "pcm16",
"temperature": 0.8,
"max_output_tokens": "inf",
"usage": {
"total_tokens": 1028,
"input_tokens": 1010,
"output_tokens": 18,
"input_token_details": {
"text_tokens": 779,
"audio_tokens": 231,
"image_tokens": 0,
"cached_tokens": 960,
"cached_tokens_details": {
"text_tokens": 768,
"audio_tokens": 192,
"image_tokens": 0
}
},
"output_token_details": {
"text_tokens": 18,
"audio_tokens": 0
}
},
"metadata": null
}
Also broke for me was working perfectly just a couple days ago
Same issue here - server error on response.done.
I believe the problem lies with the
”gpt-4o-realtime-preview-2024-10-01” model. We changed to
”gpt-4o-realtime-preview-2025-06-03” and it worked, at least the small test. You can change this on the wss for the OpenAI Realtime Model
We will try testing some more but I hope you can try in the meantime and see if this helps you 
!
Thanks for your reply, I checked it again and it worked
Looks like the problem fixed itself. Still a bit frustrating there is not a word from OpenAI about this. (or at least i can’t find it)