Best Codex Usage Trackers for Mac and Windows (2026)

How to choose

• Where should the number live? A page you open, a command you run, or an app that stays in the menu bar or taskbar.
• Which computers? CodexBar and Codex Limits are Mac apps (CodexBar also has a Linux desktop app); Windows needs one of the apps in the Windows section.
• Do you want a warning? Only some trackers notify you before a limit; the official options never do.
• One Codex account or several? Work and personal ChatGPT logins need a tracker that keeps them apart.
• Free or paid? Everything here is free except SessionWatcher. Pay only for a specific gap.

Why a message count doesn't tell you where you stand

OpenAI publishes Codex allowances as ranges of messages per usage window rather than fixed counts, and explains why. Per the Codex pricing documentation: “The number of messages you can send depends on the model used, size and complexity of your tasks, and whether you run them locally or in the cloud.” The same page adds that small scripts may consume only a fraction of an allowance, while larger projects, long-running tasks, or extended sessions that require the agent to hold more context use significantly more.

One shared allowance is documented, with a longer limit possible on top. From the plan rate tables: “Local messages and cloud chats share your plan's usage allowance. Weekly limits may also apply.” So local messages and cloud chats draw on the same allowance rather than two separate ones (counted per five-hour period on Plus and Standard Business; Pro plans currently have no five-hour limit), and a weekly limit may sit above it.

In the same table footer, as separate notes, OpenAI states that Enterprise and Edu users with flexible pricing have no fixed rate limits because usage scales with credits, and that Enterprise and Edu plans without flexible pricing have the same per-seat usage limits as Plus for most features.

Past the included allowance, the unit changes. OpenAI: “Credits are the unit used to pay for eligible usage on credit-based plans. After you reach your included limits, available credits let you continue working.” OpenAI publishes credit rates per million input, cached input, and output tokens but no dollar-per-credit conversion in that source, so we do not convert credits into dollars anywhere on this page.

Speed settings change the burn rate too. OpenAI states that you can increase model speed in Codex “in exchange for higher usage,” and that image generations use included limits roughly 3-5x faster on average than similar turns without image generation, depending on image quality and size. The Speed documentation gives the multipliers for its faster modes.

So the practical question is not “how many prompts have I sent” but “how much of this window have I spent, and when does it reset.” Below are the places that answer some part of that.

The Codex usage dashboard (official)

OpenAI's answer to “where can I see my current usage limits” is the Codex usage dashboard at chatgpt.com/codex/settings/usage. From the pricing documentation: “You can find your current limits in the usage dashboard.” That is where to find your current limits and reset times. It does not document a live reading, and we make no claim about how quickly the dashboard reflects a session either way.

Price: included with the plan. Limitations: it is a page you open, so it is not visible while you work, and no notifications are documented. If you use Codex with an API key instead of a ChatGPT plan, OpenAI documents that you pay based on API pricing, so your API billing view is the one to reconcile against.

Codex CLI /status (official)

The in-terminal option, documented in the same passage: “If you want to see your remaining limits during an active Codex CLI session, you can use /status.” OpenAI's Codex CLI documentation lists it among the slash commands. Price: free. Limitations: it is a command you run during a session, so it tells you nothing when the CLI is closed, and it sends no notifications.

CodexBar (free, Mac and Linux)

CodexBar, by Peter Steinberger, is an MIT-licensed menu bar app for macOS 14 or later with about 22,000 GitHub stars, a Linux desktop app and a CLI. Its README lists 91 providers. For Codex, its docs describe session and weekly lanes with reset countdowns, credits, local cost scans, optional session quota notifications, and the list of banked resets (“reset credits”) with nearing-expiry notifications; it does not redeem them. It installs with brew install --cask codexbar. Its author ships no Windows build. If you are on a Mac and want it free, start here; our SessionWatcher vs CodexBar page says the same.

Codex Limits (free, Mac)

Codex Limits is a smaller MIT-licensed menu bar app built around Codex pacing: weekly usage remaining as the main figure, the reset time, a suggested pace and runway, banked resets with a reminder 24 hours before the next one expires, and the local tasks the Mac observed. Version 0.3 adds opt-in Claude Code (experimental) and Grok (beta). It needs a signed-in Codex CLI, keeps each Codex account separate, and its README says the stable release is a universal app for Apple Silicon and Intel that is not Developer ID signed or notarized. It describes itself as an independent, unofficial project.

ccusage (free, any terminal)

ccusage is an MIT-licensed CLI that “reads local usage data from coding agent CLIs and turns it into daily, weekly, monthly, and session reports.” Its README lists 18 agent CLIs, Codex among them with the command ccusage codex daily, alongside Claude Code, OpenCode, Amp, Droid, Codebuff, Hermes Agent, pi-agent, Goose, OpenClaw, Kilo, Kimi, Qwen, GitHub Copilot CLI, Gemini CLI, Antigravity, Grok Build CLI and ZCode. Run it with npx ccusage@latest; the latest release on October 6, 2026 was v20.0.26 (September 27, 2026).

Limitations: it reports on data you have already generated, so it does not warn you live. Its five-hour blocks view is documented for Claude Code, not for Codex windows. Claude-Code-Usage-Monitor, the other popular terminal monitor, documents no Codex source.

Tracking Codex usage on Windows

CodexBar has no Windows build, but several free apps do the job, each checked October 6, 2026 against its README:

• Claude Code Usage Monitor by CodeZeno: a taskbar widget for Claude Code, Codex, Antigravity, OpenCode Go, Cursor, Grok Build and Copilot, with several Codex accounts; install with winget.
• Win-CodexBar: a community port of CodexBar to the system tray, 56 providers per its README; winget package Finesssee.Win-CodexBar.
• OpenQuota: a tray app for Windows, macOS and Linux with Codex session and weekly limits, credits and token history; installers not code-signed.
• SessionWatcher for Windows (ours, paid): the Codex reading inside the Windows 11 taskbar, or in a small window beside it on Windows 10 (version 2004 or later), with the same plans as the Mac app and one licence for both. The installer is not code-signed yet, so SmartScreen asks once.

More detail is in CodexBar for Windows and the best AI usage trackers for Windows.

SessionWatcher for Codex Our own product

SessionWatcher is paid, and the free apps above cover part of what it does. What it adds for Codex users, from our facts page:

• Mac and Windows on one licence: the menu bar on a Mac, inside the taskbar on Windows 11; Pro covers up to 5 devices in any mix.
• Several Codex accounts side by side (and Claude accounts), each with its own limits and reset times: Codex multi-account, Pro.
• Banked resets: listed with their expiry, an alert before one expires on every account you track, and on the Mac applied after you confirm. See Codex resets and banked resets.
• Warnings at 70% and 90% of a limit by default, or every 10%, and when it resets.
• History and cost: 7, 30 or 90 days of daily usage, and a token table built from Codex's local session logs with an estimated cost. Your billing view settles what you actually pay.
• iPhone and Apple Watch companion in TestFlight beta, synced from your Mac.
• Support from the developer by email, and a 14-day refund.

It tracks 20 tools; CodexBar lists more. It has no Linux build and no CLI. macOS 14+ or Windows 10 (2004+)/11 only.

Comparison: what each source documents

Sources for this table, all read October 6, 2026: OpenAI's Codex pricing documentation and Codex CLI documentation; the CodexBar, Codex Limits and ccusage READMEs and docs. SessionWatcher cells come from our own facts page and were not independently tested.

Frequently Asked Questions

See also our how to check Codex usage guide for every manual way to look.