How to check your Codex usage
Codex ships with generous limits on ChatGPT plans — but they are limits, and they arrive mid-task if you're not watching. Here's how to see how much you've used, and the fastest way to keep the number in front of you while you work.
What counts toward your Codex limit
Codex usage on ChatGPT plans is metered on a rolling window of about five hours, with a weekly cap on top. Every prompt you send counts, and so does the work the agent does on your behalf — long autonomous runs consume more of the window than quick edits. The CLI warns you as you approach the cap, but like most rate-limit warnings, it shows up when you're already close to the edge, not when you're planning the next task.
Three ways to check your usage
Run /usage in the Codex CLI
Type /usage in a Codex session to open the usage
view for your account. /status complements it with
the current session's configuration and token usage, including
how much context capacity remains. Accurate and official — but
it's a snapshot inside one session, and you have to remember to
ask.
Check your ChatGPT plan
Your ChatGPT account shows plan-level information in the browser. It's fine for reviewing what your subscription includes, but it's several clicks away and not something you'll consult in the middle of a coding session.
Use a notch or menu-bar monitor
A lightweight Mac app that reads your Codex session can keep the current window percentage and reset countdown permanently visible. It's the only option that answers "can I start one more big run?" before you commit to it.
The effortless way: your usage in the notch
Vibe Island reads your existing Codex login and shows the live numbers in your Mac's notch: the percentage of the current window you've used and the time until it resets, for both the rolling window and the weekly cap. No command, no browser tab, no API key. Kick off a long agent run, watch something else entirely, and the moment the question becomes "do I have room for another one?" — the answer is already at the top of your screen.
How the two windows work
Key detail
The rolling window and the weekly cap run at the same time — whichever is closer to its limit is the one that will stop you.
The rolling window resets a few hours after you start using it, on its own schedule rather than a fixed clock; the weekly cap refills on a longer cycle. That's why a single "usage" number is misleading — what you actually want to know is which window is tightest right now and when it resets, which is exactly what a live countdown gives you. More on the usage tracking page.
What to do when you hit the limit
Options, in rough order of effort:
Wait it out. The rolling window often resets sooner than you'd expect — with the countdown visible, you know whether it's worth waiting or worth switching.
Use a lighter model. Heavier reasoning modes consume the window faster; routine refactors and reviews usually don't need them.
Route work to another agent. Claude Code, Gemini CLI, and other agents have entirely separate budgets. If you monitor all of them in one place, you can see who has headroom before you switch instead of guessing.
Upgrade the plan. If you're hitting the wall every day, the higher ChatGPT tiers come with substantially more Codex usage.
Common questions
How do I see how much Codex usage I have left?
Run /usage inside the Codex CLI to open the usage view for your account, or /status for the current session's token usage and remaining context. For a number you never have to ask for, a notch monitor like Vibe Island keeps the percentage used and the reset countdown visible at the top of your screen.
What is the Codex 5-hour limit?
Codex included with ChatGPT plans is metered on a rolling window of roughly five hours, with a weekly cap on top. Messages and the work the agent does count toward the window; when you reach the cap, Codex pauses until the window resets.
Do I need an API key to check my Codex usage?
No. If you use Codex through your ChatGPT plan, your usage is tied to the account you're already signed into. Vibe Island reads your existing CLI session — no API key and no extra dashboard.
Can I see Claude Code usage in the same place?
Yes. The same notch panel tracks usage for Claude Code, Kimi, GLM, and DeepSeek alongside Codex, each with its own windows and reset times — useful when you route work to whichever agent has headroom.
Every agent's usage in one place
Codex is rarely the only agent on a Mac. Vibe Island tracks Claude Code, ZCode, Gemini CLI, Antigravity CLI, Cursor Agent, Trae, OpenCode, MiMoCode, Factory Droid, Qoder, Qwen, Grok Build, Kimi Code, DeepSeek, Mistral Vibe, Copilot, CodeBuddy, WorkBuddy, Kiro, Hermes, Amp, Pi Agent, Oh My Pi, and Gajae Code. in the same notch panel — sessions, approvals, and usage together. See the full usage breakdown, read how Codex rate limits work, or start with Codex monitoring.
Vibe Island · macOS 14+
See your Codex usage from your Mac notch
Download Free Trial2-day free trial · all features included · no card required