Codex usage limits depend on your ChatGPT plan, the model you choose and how large each task is, so OpenAI publishes estimated ranges per five-hour window rather than a fixed message count.
As of November 2, 2026, OpenAI’s Codex pricing documentation lists these estimates for local messages every five hours on Plus and Standard Business:
- GPT-6 Luna: 350 to 3,000
- GPT-6.1 Terra: 35 to 285
- GPT-6.1 Sol: 15 to 160
- GPT-6 Astra: 5 to 45
Pro plans ($100, $200 or $500 a month) currently have no five-hour limit, though weekly limits may apply on any plan. Free and Go ($8) include GPT-6 Luna in the desktop app, subject to rollout. Cloud tasks, Fast mode (2.5x) and image generation use your allowance faster. You can check remaining usage on the Codex usage dashboard or with /status in the CLI. If you hit the limit, Plus and Pro users can buy credits, or run local chats with an API key at standard API rates.
Related: Is Codex free?, Codex CLI 0.166 power-user guide and Claude Code usage limits
Source: OpenAI, “Codex Pricing” documentation
Image: ChatGPT logo by OpenAI, public domain (trademarked), via Wikimedia Commons. ChatGPT and Codex are trademarks of OpenAI. FreeAnswers is not affiliated with or endorsed by OpenAI.
2 thoughts on “Codex usage limits: how much can you use on each plan?”