Code_interpreter Generates and Image and API Assistant Fails

API Bugs
1

You can even do this in the playground. But I first found it on the API.

  1. Have code_interpreter generate any image.
  2. The assistant message will have a type of “image”
  3. Ask anything else on that same thread and the API will return

{
“error”: {
“message”: “Invalid model: gpt-4-turbo-2024-04-09 does not support image message content types.”,
“type”: “invalid_request_error”,
“param”: “model”,
“code”: “invalid_type”
}
}

This needs to be fixed ASAP.

2 Likes