GPT 5.5 Codex Quota Drains Quickly

Take a look at my comment here on how to fix: Why is token usage in Codex increasing so quickly now?