For days I was wondering if I was doing something wrong; My GPT-4 requests were taking extraordinarily long or just timing out completely. Turns out, I'm not the only one experiencing this. With OpenAI's massive growth, their servers have been struggling to keep up (understandably so!), but unfortunately, their official status page doesn't give any numbers of what kind of performance we can expect, so that's why I made GPT Status. It's a super basic dashboard that I put together over the last two days. It's completely open source, and it's built with Cloudflare Workers, NextJS 13, Shadcn UI, Vercel & Vercel Postgres. I'd love to see this get improved on over time with more models being measured (Anthropic, Google, Cohere, etc), as well as scraping the official status page for uptime and incidents info. Anyways, let me know your thoughts, and if you're interested, please, please feel free to submit a PR to make it better. 😉

GPT Status

Open-source dashboard for monitoring OpenAI API performance

Open-source dashboard for monitoring OpenAI API performance