When structuring the output of function calls, there is Chinese character encoding issue resulting in garbled text

Garbled characters when outputting Chinese.

Output results:“name”: “æ\u0080»è\u0083\u0086å\u009bºé\u0086\u0087”。

Correct result:“name”: “总胆固醇”。

Everything is normal except for the latest model.

I’m currently using gpt-3.5-turbo-1106 and having the same issue, with garbled characters in function calling response.

Sample: “å<81><9c>è»<8a>å ´å<96>ºé<82><8a>”

It goes back to normal when I switch to gpt-3.5-turbo-0613.

1 Like