AI Token Counter for GPT, Claude & Gemini

Count tokens for GPT, Claude and Gemini and estimate the API cost — paste your text and see the numbers update instantly, all in your browser.

This token counter tells you how many tokens your text uses for GPT, Claude and Gemini, and estimates the API cost. Paste a prompt to count tokens for ChatGPT, check whether it fits a model’s context window, or compare token usage across models before you send it. The number updates as you type, with words and characters shown alongside. Nothing is uploaded.

Read the guide: How to Count Tokens for GPT, Claude & Gemini
Counts update as you type — nothing is sent anywhere.
Your text or prompt
Count & cost
0tokens

Exact for GPT-4o (OpenAI).

15
Words
81
Characters
1
Sentences
Context window0 / 128,000

Estimated cost

Input (0 tokens)$0.00
Output (500 tokens)$0.00500
Total per request$0.00500
Prices are editable estimates. Verify the current rate on the provider’s official pricing page before relying on it — batching and prompt caching can lower cost.

How it works

  1. 1

    Paste your text

    Type or paste the prompt or document you want to measure into the box.

  2. 2

    Pick a model

    Choose a GPT, Claude or Gemini model. The token count updates instantly for that model.

  3. 3

    Read count and cost

    See tokens, words and characters, whether it fits the context window, and an editable cost estimate.

Instant & 100% private — nothing is uploaded

Everything runs locally in your browser. Your code, text and files are processed on your own device and are never sent to a server — so there are no upload waits, no size limits from us, and nothing is ever stored or logged.

Frequently asked questions

How many tokens is 1,000 words?
Roughly 1,300 to 1,500 tokens in English, because one token is about 0.75 words. Code and non-English text usually run higher.
How do I count tokens for ChatGPT?
Paste your text and select a GPT model. It shows the exact token count for that model’s tokenizer instantly, with no API call.
Is the token count accurate for GPT-4o?
Yes. For OpenAI models it applies the same tokenization the API bills against, so the count matches what you are charged. Claude and Gemini counts are close estimates because each uses its own scheme.
Are tokens the same for Claude, GPT and Gemini?
No. Each model family tokenizes text differently, so the same sentence can produce a different token count on each. Always count against the model you will actually use.
How much do tokens cost?
It depends on the model and whether tokens are input or output, with output usually priced several times higher. The tool shows an editable estimate; check the provider’s pricing page for the current rate.
Is my text sent to a server?
No. All counting happens in your browser on your device. Your text never leaves your computer and nothing is uploaded or stored.

More tools

JSON Formatter & ValidatorBeautify, minify, validate JSON with a collapsible tree view.Code Formatter & MinifierPretty-print or minify JS, TS, CSS, HTML, JSON, SQL and more.YAML ⇄ JSON ⇄ TOML ConverterConvert config between YAML, JSON and TOML, both ways.JSON Schema Viewer & ValidatorRender a JSON Schema as a tree and validate data against it.Markdown Editor (Live Preview)Write Markdown with a live, side-by-side HTML preview.Mermaid Live EditorTurn Mermaid text into flowcharts and diagrams, live.

More from the Hivly network

Free sister tools on our other sites.

Color Palette GeneratorShuffle palettes with the spacebar, lock favourites, copy hex codes.Length ConverterMetres, feet, miles, inches, km and more, instantly.Word CounterCount words, characters, sentences and reading time live.