search

kydycode / chatgpt-3.5-turbo

A Python chatbot that uses OpenAI's GPT-3.5-turbo model to generate responses to user input. Simply enter your input and the chatbot will respond. Conversation history is stored in a list. Licensed under MIT.
167 stars 71 forks source link

openai.error.InvalidRequestError #2

Open myrzx opened 1 year ago

myrzx commented 1 year ago

You: Hi, are you a chatbot for me? Traceback (most recent call last): File "C:\Users\Otp_Lab\Desktop\LXH2022\Fun\chat.py", line 77, in main() File "C:\Users\Otp_Lab\Desktop\LXH2022\Fun\chat.py", line 48, in main response = send_message(message_log) File "C:\Users\Otp_Lab\Desktop\LXH2022\Fun\chat.py", line 10, in send_message response = openai.ChatCompletion.create( File "E:\Python\lib\site-packages\openai\api_resources\chat_completion.py", line 25, in create return super().create(*args, **kwargs) File "E:\Python\lib\site-packages\openai\api_resources\abstract\engine_apiresource.py", line 153, in create response, , api_key = requestor.request( File "E:\Python\lib\site-packages\openai\api_requestor.py", line 226, in request resp, got_stream = self._interpret_response(result, stream) File "E:\Python\lib\site-packages\openai\api_requestor.py", line 619, in _interpret_response self._interpret_response_line( File "E:\Python\lib\site-packages\openai\api_requestor.py", line 679, in _interpret_response_line raise self.handle_error_response( openai.error.InvalidRequestError: This model's maximum context length is 4096 tokens. However, you requested 4135 tokens (39 in the messages, 4096 in the completion). Please reduce the length of the messages or completion.

kydycode commented 1 year ago

Please reduce the length of the messages or completion.

myrzx commented 1 year ago

Please reduce the length of the messages or completion.

It's the first request and I just said one sentence. I change the max_tokens=4096 to 2048, it run. What is '4096 in the completion' mean? 4096 empty tokens in the message to chatgpt api?

jane00 commented 1 year ago

same question

myrzx commented 1 year ago

same question

From OpenAI Document max_tokens The maximum number of tokens allowed for the generated answer. By default, the number of tokens the model can return will be (4096 - prompt tokens). Prompt tokens (your message) + max_tokens should <= 4096, else it raise error.

kydycode commented 1 year ago

That flow works like that - previous messages are added to each subsequent request so that the context of the conversation is kept. so after a while, the number of tokens can get to more than 4096

myrzx commented 1 year ago

That flow works like that - previous messages are added to each subsequent request so that the context of the conversation is kept. so after a while, the number of tokens can get to more than 4096

https://platform.openai.com/docs/api-reference/chat/create#chat/create-max_tokens max_tokens defaults to inf, if set 3800, the send messages must < 296 tokens, it will raises error quickly.