Compared

Notchlet vs ccusage

ccusage is a report, Notchlet is a gauge. ccusage tells you what you spent, in detail, for more agents than Notchlet reads. It cannot tell you how much of the 5 hour window is left, because the logs do not know. Notchlet asks the provider for that and keeps a smaller history next to it.

Side by side. Checked Sep 4, 2026.

 Notchletccusage
Lives inThe notch. Nothing until you hover.Your terminal. Run npx ccusage@latest.
TracksClaude Code, Codex, Cursor and OpenCode, three at a time.Over 18 agent CLIs, from Claude Code and Codex to Gemini CLI and Copilot CLI.
ReadsThe login each CLI already saved, then the usage endpoint behind /usage or /status. Never browser cookies.The local usage logs each CLI writes. No network needed.
History and costCost and tokens for today, 7 and 30 days, a 12 month activity graph and per-model tokens, from the logs on your Mac.Daily, weekly, monthly, session and 5 hour block reports with cost per model. Remaining rate-limit windows are not part of it.
ShareA PNG of your numbers, graph and models.JSON export.
PriceFree.Free.
SourceMIT on GitHub.MIT on GitHub.
Runs onmacOS 14 or later.Anywhere Node runs.

ccusage facts from its README. If one is wrong, open an issue and it gets fixed.

Which one. Pick by what you run.

ccusage if

  • You want a detailed report you can script or export.
  • You use agents beyond Claude Code, Codex, Cursor and OpenCode.
  • You are not on a Mac.

Notchlet if

  • You want to know what is left, not only what you spent.
  • You want the number visible without a terminal command.
  • You want history that survives Claude Code deleting its transcripts after 30 days.

Stop finding out you are rate limited the hard way.

Download for macOS

Free and open source. macOS 14 or later.