You've burned money on AI API calls without knowing it. Everyone who builds with Claude, GPT, or Gemini has.
Here's the pattern: you write a prompt, run it a few hundred times while testing, and only find out what it actually cost when the invoice lands at the end of the month. By then it's too late to change anything.
I built ClearBurn to fix the timing. It shows the cost while you're still writing the prompt, not after you've already sent it.
What it does
- Live cost, inline, the moment you write a prompt. No commands, no tabs, no calculator.
- Every model priced side by side (Claude, GPT, Gemini), so you can pick the cheapest one that still does the job.
- A full project scan that finds every place you're calling an LLM API and projects what it'll actually cost at real volume, not just once.
- CSV export, ready for an expense report or a budget conversation with your team.
Free vs Pro
The core extension is free on the VS Code Marketplace install it, and you get live inline cost estimates right away.
ClearBurn Pro adds full project scanning, cross-model cost comparison, and CSV export. It's $9, one time not a subscription, not a trial that nags you.
Setup takes under a minute: install the free extension, paste in your license key if you go Pro, done.
Try it
Install the free extension from the https://marketplace.visualstudio.com/items?itemName=artcllick-apps.clearburn or straight from the Extensions tab in VS Code.
If you find it useful, a review on the Marketplace would mean a lot it's a small team and every bit of feedback helps.
Happy to answer questions in the comments what would make this more useful for your workflow?
Top comments (0)