Token Counter
Count tokens and see how text is split, with the real GPT-4o and GPT-4 tokenizers.
Tokens
Examples
Command output
The RTK Output Visualizer counts the tokens in git, ls, grep and test output before and after rtk-style filtering.
Questions
Which models does it count for?
OpenAI models that use o200k_base (GPT-4o, GPT-4.1 and the o-series) or cl100k_base (GPT-4 and GPT-3.5). Claude, Gemini and Llama use their own tokenizers, so their counts differ, often by 10 to 30 percent.
Why does an emoji or Chinese character take several tokens?
Tokenizers work on UTF-8 bytes. A common English word is often one token, but rarer characters that take several bytes are split into pieces. The highlight groups the pieces back into the character.
Is my text uploaded?
No. The tokenizer data (0.5 to 1 MB) downloads once from this site, and counting runs in your browser.
Related tools
- RTK Output VisualizerSee how rtk-style filtering shrinks command output before an AI agent reads it.
- LLM Sampling ExplorerSee what temperature, top-k, top-p and min-p do to a model's next-token choices.
- LLM Memory CalculatorHow much GPU memory a model needs: weights plus KV cache, with the maths shown.
- Word & Character CounterWords, lines and three kinds of character count.