I used $30,983 of AI tokens last month in Claude code on $200/mo plan
I built tokenflex.ing — a public leaderboard where devs and vibe coders can show off how much ai tokens they have used with Claude Code, Codex, OpenCode, Cursor and other AI tools.
Kind of like a GitHub profile, but for AI token usage.
My profile so far:
- 18.7B tokens all-time
- 17B tokens last 30 days
- 51,414 Claude Code events
- 5 models (Opus 4.6, Opus 4.7, Sonnet 4.6, Codex, synthetic)
- Currently rank #4 today
Funny thing is nobody actually knows their actual token usage until they look. You assume you're using "a lot" of Claude. Then you see 12B tokens for the
month and have to quietly close the tab.
What does yours look like?
Especially if you're curious about anyone running Claude Code + Cursor + Codex together you must compare your usage with them.
You should try tokenflex.ing if you want to do that.
The idea is interesting because token usage is becoming a real identity/status signal for AI-heavy developers, not just a billing metric.
The part I’d pressure-test is the name. Tokenflex is funny and memorable, but it also makes the product feel like a novelty leaderboard. If this stays as a fun flex page, that works. But if you want it to become a serious profile layer for AI-native developers across Claude Code, Cursor, Codex, OpenCode, and future tools, the name may cap how seriously people read it.
Exirra .com would fit that bigger direction better because it feels more like an AI usage intelligence/profile layer, not just a token bragging board. The product already has the raw material for something bigger: usage identity, model comparison, developer credibility, and public AI-native work signals.
I’d decide that before more profiles and sharing loops build around Tokenflex, because the name will shape whether people see it as a meme or an actual developer reputation layer.