Gpt repeatley respond wrong api code(chat.completions)

Hello, I am a user of the ChatGPT API. I am writing code through GPT, but the OpenAI API is continuously being used incorrectly. It keeps using the old openai.ChatCompletion.create, even though I correct it to chat.completions.create in the chat or provide the API documentation page. Despite sending dozens of messages, it occasionally uses the correct input but continues to revert to openai.ChatCompletion.create. I don’t understand why this is happening. Is there any solution to this problem?

1 Like

This is the most intentional and annoying thing ever. If you use a chat.completion it always reverts it to trash. I know it’s dated but feed it its own user manual for the love of God.

1 Like

Yes, came here for this reason, and it doesn’t matter if you’re using the latest ChatGPT models. When you try to correct ChatGPT, it will even tell you that you’re wrong. :joy: