Token Counter
Estimate tokens, characters, and context usage for your prompts. All counting happens locally.
Estimates only — actual tokenizer counts may differ. Nothing is sent to a server.
Are the token counts exact?
They are close estimates. Exact counts depend on each provider's tokenizer, but the estimate is reliable for planning.
Is my text private?
Yes. Counting happens entirely in your browser, so your prompts stay on your device.
Which models are supported?
Estimates cover popular families such as GPT, Claude, and Gemini.