Skip to main content
Files

In this section of CodeGPT's Studio you'll see a list with all the Files you've already uploaded (If you have) or upload your fist one!

Updated over 2 months ago

How to upload a File

  • Click on Upload.

  • Similar to when you upload a File directly to an Agent, this window will pop up:

  • Here you can select the type of File you wish to upload:
    -From your desktop.
    -Create one from scratch.
    -By an URL.

  • Once you've selected the one you want to upload, confirm by clicking on Import

  • Refine the Data, here you can add or delete text, change the name of the file. When you're done click on Start Training.

  • Now it will appear in the list as the other uploaded files and it's ready to be uploaded to your agents! Visit this link to learn how.


โš ๏ธ File Size and Storage Capacity Information

Depending on your plan, there's a maximum file storage limit to consider. The maximum file upload size is 20MB, but the resulting file size after the vectorization process can vary based on several factors, such as:

  • Language Model Used: Different models like GPT-4o have their own tokenization strategies, meaning the number of tokens can differ depending on the model.

  • Text Length: The number of tokens needed depends on the length of the text. You can estimate the token count by dividing your text into tokens. Keep in mind that some models have limits on the maximum number of tokens they can process in a single call.

Example:

Using the GPT-4o model, the following sentence would be divided into approximately 22 tokens:

"The artificial intelligence is transforming many aspects of our daily lives. From virtual assistants to recommendation systems, AI is present in various applications."

GPT-4o, like its predecessors, tokenizes text into tokens which are smaller segments of text rather than just words.

To estimate the memory used:

  • Assume each token occupies approximately 4 bytes.

  • The conversion for this example would be:

22 tokens * 4 bytes/token = 88 bytes

Thus, the longer the text, the more bytes the resulting vector will occupy.

For more information about tokens and how to estimate them, you can use this app


How did we do? If you want, you can give us your review here: CodeGPT

Join our Discord Community here!: Join the CodeGPT Discord Server!


If you have an issue, feel free to document it in: Issues ยท JudiniLabs/code-gpt-docs
Or contact us if you need further assistance!

Did this answer your question?