TokenPrint

compare AI coding usage, estimated API costs, and environmental impact

TokenPrint brings usage from Claude Code, Codex CLI, and Gemini CLI into one local dashboard. It shows token counts over time alongside estimated API costs, energy use, carbon emissions, and water use.

Run locally

The Python CLI installs from the repository at github.com/sburl/TokenPrint:

git clone https://github.com/sburl/TokenPrint.git
cd TokenPrint
python3 -m venv .venv
source .venv/bin/activate
python -m pip install .

Then follow the repository’s collector setup for the tools you use. Claude Code and Codex use usage collector commands; Gemini uses local telemetry logs. Once those are configured, tokenprint --no-open --output report.html generates the report.

The project also has an optional Go daemon for a dashboard that refreshes over time. The instructions above use the one-shot Python CLI.


Standing invitation (inspired by Patio11 who also has some good tips on how to approach this): if you want to talk about hard tech or systems, I want to talk to you.

My email is my full name at gmail.com.