X · @simonw
· X / Twitter
GPT-5.6 found optimizations that "reduced end-to-end serving costs by 20%" for OpenAI to serve that model Presumably that's billions of dollars a mont…
GPT-5.6 found optimizations that "reduced end-to-end serving costs by 20%" for OpenAI to serve that modelPresumably that's billions of dollars a month in savings at this point?Vaibhav (VB) Srivastav: Codex analysed production traffic, improved load balancing, rewrote production GPU kernels and ran hundreds of experimen