GPT-6 Sol feels like a minor upgrade and burns usage fast, a paying Codex user says
A paying Codex user compares Opus 5.5, Astra and GPT-6 Sol, saying Sol is only cheaper rather than a real upgrade and that usage runs out too quickly.
There was a time when Codex was great at everything. Codex had a strong model and fair usage back in the GPT 5.5 and 5.6 days. Nobody was complaining about limits. Now I'm seeing the same thing with Opus 5.5 because it's fast, really good at coding and the usage actually lasts without burning through my limit. Astra is different because it's smart but painfully slow and eats usage so fast no matter which thinking level I pick, so the limit is gone by the end and even after the usage resets it feels like it runs out faster than before. And GPT 6 Sol hasn't impressed me either because it doesn't feel like a big upgrade from 5.6 Sol, it's just cheaper and that's it. I'm paying for Codex and still running out of usage. Unless OpenAI brings a GPT 6.1 Sol or Astra that's smarter and has better limits, I don't see a reason to stay.