Скачать или смотреть Optimizing GPT4 Token Limits: Understanding how to Efficiently Manage Prompts

The tools I develop are available on https://bionichaos.com
You can support my work on   / bionichaos  

I'm developing locally and have issues with the initial window size when the page loads, indicating a frontend script problem. I need to examine the code to identify and address it.

There's confusion about the prompt limit for GPT-4 users. Currently, there's an hour timer translating to one prompt every 3.6 minutes over 180 minutes. A sliding time window with shorter intervals might be more efficient. Switching to GPT-3.5 prompts doesn't seem to revert to GPT-4 based on past tests.

Prompt limits are calculated based on token consumption. For GPT-3, the token limit, including both input and output, is approximately 4096 tokens. Tokens are units of text, from a character to a word. The total tokens, combining input and response, shouldn't exceed the model's token limit. Crafting concise prompts can optimize token use.

A suggestion is to use GPT-3 for creating shorter prompts for GPT-4 to manage token limits and reduce timeouts, which seems innovative but depends on the tools' integration capabilities.