Using OpenAI API

GPT-4o

Simon Willison notes

The size of the vocabulary of the tokenizer - effectively the number of unique integers used to represent text - has increased to ~200,000 from ~100,000 for GPT-4 and GPT-3:5. Inputs in Gujarati use 4.4x fewer tokens, Japanese uses 1.4x fewer, Spanish uses 1.1x fewer. Previously languages other than English paid a material penalty in terms of how much text could fit into a prompt, it’s good to see that effect being reduced.

Developer Survey

Anatomy of OpenAI’s Developer Community is an extensive analysis (with Python code) of all the issues and discussions held on OpenAI’s developer forums.

For details see atlas.nomic.ai

Tips

Turn off OpenAI scraping for your website with these robots.txt additions.

User-agent: GPTBot
Allow: /directory-1/
Disallow: /directory-2/

Also see How to build your own OpenAI cost calculator

Eden AI is a API for APIs that lets you pick and choose from among multiple generative, text analysis, image generation and other APIs.

ReliableGPT is a Python library that optimizes the OpenAI API.

Plugins

OpenAI Examples

See Tweets from OpenAI developer relations @OfficialLoganK

Github: Chat Retrieval

a new plugin which allows uses to host their own data and make it accessible inside of ChatGPT…

Libraries

simpleaichat a Python package for easily interfacing with chat apps like ChatGPT and GPT-4 with robust features and minimal code complexity.

Tracking the API

OpenAI Cookbook a Github repo of examples of using the OpenAI API.

Unofficial dashboard with daily updates of the speed of OpenAI API calls:

from openai-status.llm-utils.org