Hey @embirico
After gpt-5.1-max launched, codex has been strangely running slower (2-3 minutes for basic responses) than in the prior codex version (gpt-5 is also slow now). I run codex through a Docker environment on Windows for Linux. Any idea why?
@haydendevs The script it wrote for me corrupted the file and then it encouraged me to get backups from my colleagues that don’t exist when the backup stored in git wasn’t sufficient.
@NoahEpstein_ Your spot on witherror handling. Bespoke builds built for custom business use cases will always have errors. The response time in dealing with them is part of what the client is paying for.
Even if you have a low error rate, not being aware of the errors that exist is an issue