Limits guide
Codex usage limits: the 5-hour and weekly windows
How Codex usage limits work on ChatGPT Plus, Pro, and Business, how many messages fit in a 5-hour window, and how to check what is left.
· Facts checked October 3, 2026
- Codex
The short answer
Codex on a ChatGPT plan draws from a five-hour usage window, and weekly limits may apply on top of it. How many messages fit depends on the model: OpenAI estimates 15 to 150 local messages per five hours with GPT-6 Sol on Plus, and 350 to 3,000 with GPT-6 Luna.
At a glance
| Plans with Codex | Plus ($20 a month), Pro ($100, $200, or $500), Business, Edu, and Enterprise. Free and Go have limited access. |
|---|---|
| Five-hour window | Plus and Business share the same estimates. OpenAI says Pro plans currently have no five-hour limit. |
| Weekly limits | OpenAI says weekly limits may also apply. |
| What counts | Local messages and cloud tasks share one allowance; cloud tasks can use more of it. |
| Where to check | /status in the Codex CLI, or the Codex usage page in ChatGPT settings. |
| Past the limit | Buy credits on Plus or Pro, or run local tasks with an API key at API rates. |
What a five-hour window holds
OpenAI publishes ranges rather than fixed counts, because a message that reads a large codebase or runs for a long time uses more of the allowance than a short edit. On Plus and Standard Business, the estimates per five hours are 5 to 45 local messages with GPT-6 Astra, 15 to 150 with GPT-6 Sol, and 350 to 3,000 with GPT-6 Luna.
The model you pick is therefore the biggest lever. Running routine work on GPT-6 Luna and saving GPT-6 Astra for the hard problems stretches the same window much further. Fast modes draw on the allowance faster as well: OpenAI lists fast mode at 2.5 times the usual rate.
Weekly limits and the Pro tiers
OpenAI says Pro plans currently have no five-hour limit, and that weekly limits may also apply across plans. It does not publish weekly figures, so the practical way to plan around them is to watch the remaining percentage and its reset time rather than count messages.
Codex shares its limits and credits with ChatGPT Work. Heavy agent use elsewhere in ChatGPT reduces what is left for Codex in the same window.
Checking what is left
Type /status in the Codex CLI to see the current limits, or open the Codex usage page in ChatGPT settings. The CLI can also show rate limits in its footer through /statusline.
everyaitoken reads the same limits from your installed Codex, using the ChatGPT sign-in Codex already has, and refreshes them on a schedule you choose, from every minute to every 15 minutes. If you use a personal and a work account, each one gets its own limits. See how Codex compares with Claude Code in the Claude Code limits guide.
Your own numbers
See every limit and its reset time.
everyaitoken reads Claude Code, Codex, Cursor, OpenRouter, Gemini CLI, and OpenCode on your Mac. It shows limits with reset times, API-equivalent cost, and what caching saved or cost, in the menu bar and 17 widgets. $9 once.
FAQ
Questions
How many Codex messages do I get on ChatGPT Plus?
OpenAI estimates 15 to 150 local messages per five hours with GPT-6 Sol, 5 to 45 with GPT-6 Astra, and 350 to 3,000 with GPT-6 Luna. The range is wide because larger tasks use more of the allowance.
Does ChatGPT Pro have a Codex limit?
OpenAI says Pro plans currently have no five-hour limit, but weekly limits may still apply. The Pro tiers differ in how much usage they include.
What happens when I hit the Codex limit?
A turn already in progress may finish. After that you can wait for the window to reset, buy credits on Plus or Pro, or sign in with an API key and pay API rates for local tasks.
How do I see my remaining Codex usage?
Run /status in the Codex CLI or open the Codex usage page in ChatGPT settings. everyaitoken shows the same limits in your Mac menu bar and widgets, next to Claude Code's.
Sources
Checked October 3, 2026. Plans and limits change; the linked pages are the authority.