Codex improvement challenge day 12: Astra Pro speed and parallel agents
A user criticizes day 12 of OpenAI's Codex challenge for giving no usage numbers, while noting GPT-6 Astra Pro gets faster responses, better tool use, and parallel agents for Pro users.
🚨 Day 12 of OpenAI's Codex improvement challenge is UNDERWHELMING for users. Two things announced today: 1. Codex and Work get more effective usage through better caching, context reuse, and request routing. But OpenAI gave us NO NUMBERS on how much extra usage we're getting. 2. GPT-6 Astra Pro gets faster responses and better tool use. Pro users can also run multiple agents in parallel in ChatGPT. Here's what pisses me off: Better efficiency sounds good, but we still don't know how much it actually improves our usage. I don't expect OpenAI to ship a groundbreaking feature every day. BUT if you're promising daily improvements, show us the impact. Some progress, but usage transparency still gets zero points. Day 12: 50 points TIBO, Give us Resets as Promised. 😭