How many tokens does this prompt use?

Count the tokens of a prompt and see how many prompts fit into a model's context window. Works with the GPT, Claude, Gemini and other models on Aether API.

Count the tokens of a prompt and see how many prompts fit into a model's context window. Works with the GPT, Claude, Gemini and other models on Aether API.

What this tool answers

• The estimated token count of any text you paste
• How many copies of that text fit into a model's context window
• The share of the context window a single prompt already uses
• The rough character-to-token ratio the estimate is based on

How it calculates

• The estimate follows the common rule of thumb that one token is about four characters of English text, which is what tokenizer-free estimators use.
• The context window of each model comes from the model catalogue, so the comparison matches the limit the API actually enforces.
• The result is an estimate. Exact counts depend on the tokenizer of the model that runs the request.

Frequently asked questions

How many characters is one token?

Roughly four characters of English text per token. Code, numbers, and unusual names tokenize differently, so a precise count needs the tokenizer of the model you send to.

What is a context window?

The context window is the number of tokens a model can read in one request, including the system prompt, the conversation history, and the answer it produces.

Does the context window include the answer?

It is the total for the request. A long answer therefore uses the same budget as the prompt that asked the question.

Related tools

LLM API cost calculator · cURL to Python, Node.js or Go · Move from the OpenAI API to Aether API · Which AI model should I use? · llms.txt generator

Run it for real

Create a key and send your first request against the same models this tool prices: the model catalogue, the credit packages, and the quickstart.

Aether API — OpenAI-compatible access to 13+ providers through one API key.