Clay Icon

Understanding Token Costs for Claygent and OpenAI API Usage

Β·
Β·

I'm trying to calculate what the cost/tokens might be of using Claygent with the OpenAI API. Does Claygent consume an average/usual amount of tokens for any given prompt? If so, how many? I ran one of my prompts through OpenAI's tokenzier and found it was roughly 400 input tokens. Is this how many tokens it would cost if I ran the same prompt with Claygent or does Claygent add a flat base/a multiplier of tokens for each run? Also, for any claygent prompts I've already run, is there anyway to see the input & output tokens? Thanks for the help! Does Token consumption change based on the LLM model used?

  • Avatar of Channeled
    Channeled
    APP
    Β·
    Β·

    πŸ‘‹ Hey there! Our support team has got your message - we'll be back in touch within 24 hours (often sooner!). If you haven't already, please include the URL of your table in this thread so that we can help you as quickly as possible!

  • Avatar of Bo (.
    Bo (.
    Β·
    Β·

    Hey, thanks for reaching out! I replied in the other thread, so I’ll close this one. 😊 P.S. Next time, please try to keep your request in one message. Thanks!

  • Avatar of Channeled
    Channeled
    APP
    Β·
    Β·

    Hi Kunwarbir A.! This thread was recently closed by our Support team. If you have a moment, please share your feedback:

  • Avatar of Kunwarbir A.
    Kunwarbir A.
    Β·
    Β·

    Haha sorry about that! I tried to delete this as sson as I saw your reply come in. Thanks Bo :) I'd added this question though: Does Token consumption change based on the LLM model used?

  • Avatar of Bo (.
    Bo (.
    Β·
    Β·

    No worries, glad I could help! 😊 To answer your question, yes, token consumption can vary based on the LLM (Large Language Model) used. Different models, such as GPT-3 or GPT-4, have different token limits and might process data differently, which can affect how many tokens are used in each request. If you’re using a specific model, it’s good to check the token limits for that model, as some might be more efficient than others. Let me know if you need more details! πŸ‘

  • Avatar of Channeled
    Channeled
    APP
    Β·
    Β·

    We haven't heard back from you in a bit, so we're going to go ahead and close things out here - feel free to let us know if you still need something!