A ccusage Alternative With a Leaderboard

If you've used ccusage (by ryoppippi), you already know how good it is at pulling token counts and cost estimates from your local Claude Code, Codex, and Gemini CLI logs. The library is mature, well-documented, and widely trusted. We know, because whoburnedmore is actually built on top of ccusage — we use it as our data layer, not as a competitor to replace.

What ccusage doesn't do is show you how your numbers compare to any other developer, let you share a link to your stats, or let you follow your friends' burn rates. That's the gap whoburnedmore fills.

What does ccusage actually do?

ccusage is a CLI tool and Node.js library that reads the local usage logs that AI coding assistants write to your disk and produces a breakdown of tokens burned and estimated cost, per tool and per day. Run npx ccusage and you get output like this directly in your terminal:

$ npx ccusageReading usage logs…   Date         Tokens       Cost  2026-06-14   8,240,000    $27.40  2026-06-13   6,910,000    $23.00  2026-06-12   5,350,000    $17.80

It's fast, local-only, open source, and great for personal spend tracking. The library has solid support for Claude Code, Codex, Gemini CLI, and a growing list of other tools. whoburnedmore uses ccusage's parser internally — so the numbers you see on the leaderboard are grounded in the same rigorously maintained parser ccusage ships.

What does whoburnedmore add on top?

The simple way to put it: ccusage shows you your numbers. whoburnedmore shows you your numbers relative to everyone else — and lets you show them off. Here are the concrete additions:

1. 1

   A public leaderboard

   One run submits your totals (never code or prompts) to the shared leaderboard at whoburnedmore.com, where you're ranked by total tokens burned. Refresh anytime to see where you stand.
2. 2

   Claimable profile with X handle

   Run npx whoburnedmore --claim to pin your leaderboard entry to your X username. Now your profile page is at /d/your-handle and anyone can find it.
3. 3

   Friends boards

   Follow other developers by X handle to build a private friends board — a live mini-leaderboard of just the people you follow. Great for team burn-rate accountability.
4. 4

   Multi-tool cross-dashboard

   Your dashboard breaks totals down by tool (Claude Code, Codex, Gemini CLI, Cursor, and more) so you can see at a glance which agent is burning the most.
5. 5

   --local private mode

   Don't want to appear on the public board? Run npx whoburnedmore --local and the dashboard opens in your browser with no data leaving your machine.

How cost is computed

Like ccusage, whoburnedmore converts raw token counts to dollars using each model's public per-token price. The formula is the same under the hood:

ccusage vs whoburnedmore: feature breakdown

This isn't a replacement — it's a stack. But if you're choosing where to spend your mental model, here is the honest side-by-side:

What data leaves my machine?

The concern is understandable: AI coding logs can contain fragments of your codebase. Both ccusage and whoburnedmore are designed to avoid that. ccusage never sends anything — it is entirely local. whoburnedmore submits only the aggregated numbers: total tokens per tool, per day, and the model name. No prompts. No file paths. No code snippets.

You can verify this before committing by running npx whoburnedmore --dry-run, which prints the exact JSON payload that would be sent without actually sending it. Or just use --local indefinitely — the dashboard is fully functional in local mode. 🛡️

Who should switch, and who should stay?

Stick with ccusage standalone if you want a zero-network-egress setup, you use ccusage as a library in your own scripts, or the terminal output is all you need. ccusage is excellent software and we are happy to depend on it.

Run whoburnedmore if you want to know where you rank, share a link to your burn stats, compare burn rates with teammates, or get the multi-tool cross-dashboard in a browser rather than the terminal. The command is the same complexity:

$ npx whoburnedmore↳ scanning local usage logs via ccusage…   claude code   24.1M    $80.30   #1 tool  codex         11.8M    $32.10  gemini cli     6.4M     $8.50  total         42.3M   $120.90   → leaderboard rank: #47 of 2,840  → profile: whoburnedmore.com/d/yourhandle

For the full three-way comparison of ccusage, tokscale, and whoburnedmore — including leaderboard depth and privacy model — see ccusage vs tokscale vs whoburnedmore. If you're new to token tracking entirely, start with how to check your AI coding token usage.