Error – promt or completion lenght

  • Resolved jerincsson

    (@jerincsson)


    2 years, 6 months ago

    Hi, I am getting error code like this:

    From OpenAI: This model’s maximum context length is 4097 tokens, however you requested 5168 tokens (168 in your prompt; 5000 for the completion). Please reduce your prompt; or completion length.

    Whatever I change: prompt or number of tokens, I am getting same response. I am using Turbo or Turbo instruct model, on gpt-4 it is occuring less, although it is often too.

    What should I do?

    Tnx

Viewing 1 replies (of 1 total)
  • Plugin Support Val Meow

    (@valwa)

    2 years, 6 months ago

    Hey @jerincsson ! 👋

    This error message is not generated by AI Engine but directly originates from the OpenAI API. Currently, there is no check performed on our end at this stage of the request.

    If this was working with versions before 1.9.94, it might be due to the removal of checks related to MaxTokens (approximately 4096 tokens) in the newer versions.

    You can refer to the recommended values for each context displayed in the Chatbot AI settings section. Additionally, trying different models that offer larger context sizes might be beneficial.

    If changing the models does not affect the query parameters, it could be a result of optimization or cache-related plugins retaining old parameters. Please disable such plugins. Thank you! 😊

Viewing 1 replies (of 1 total)

The topic ‘Error – promt or completion lenght’ is closed to new replies.