Codex users warned: 1M context setting destroys cache and burns quota

Codex users reported abnormal quota consumption traced to an experimental 1M token context window setting in codex.toml, which breaks OpenAI's prompt caching. Removing the setting restores normal usage.

2026-09-27 ~ 2026-09-27 · 3 related posts

• Codex eating your quota? A 1M-token context setting is killing your cache — altryne · 2026-09-27

2 near-duplicate retellings: altryne · thursdai_pod