Token Counter

About the Token Counter

Estimates how many tokens a piece of text will use for large language models. It uses the common heuristic of about 4 characters (≈0.75 words) per token for English.

How to use

Paste your text, then click Count.

Worked example

A 100-word paragraph ≈ 130 tokens.

How to use this calculator

1. Paste or type your text into the tool.
2. Review the instant estimated token count.
3. Use the result to compare against your model's context limit or prompt budget.

The formula explained

• \(T\) = estimated number of tokens
• \(x\) = the input text
• \(\text{token count}(x)\) = a tokenizer-dependent estimate of how many tokens the text contains

Step by step method

1. Start with the exact text you want to send to an LLM.
2. Enter it into the Token Counter.
3. Read the estimated token total shown by the tool.
4. If needed, shorten or split the text and check again.

Worked example

Suppose you want to send a short update to an LLM and need to see whether it is likely to fit within your prompt budget.

1. You paste this text: "Please summarize the meeting notes and highlight the action items.".
2. The tool estimates the text at 10 tokens.
3. You compare that estimate with your available limit and decide whether you can include more context.

Tips and common mistakes

• Token counts are estimates, not exact promises, because different models can tokenize text differently.
• Short words, punctuation, and spacing can affect the count, so always check the full text you plan to use.
• If your prompt is close to the limit, leave room for the model's reply as well as your input.
• For long documents, split the text into chunks and count each chunk separately.

Frequently asked questions