A token is the unit an LLM bills and reads. In English, one token is about 4 characters or about three quarters of a word. Other languages and code often use more tokens for the same text.

Why do counts differ between models?

Each family has its own encoding. This tool counts OpenAI text with o200k_base and can show cl100k_base. Other providers are not the same encoding.

Are Claude and Gemini counts exact?

No. Claude and Gemini do not publish a browser tokenizer. Their counts are an estimate equal to the o200k_base count. Use the provider's count-tokens API when you need an exact number.

How is cost calculated?

Input and output are priced separately, per million tokens. Cached input uses the cache-read rate when the model has one. A request over a model's long-context threshold uses that model's higher rate for the whole request. Monthly cost is the request cost times requests per day times 30.

No. Counting and the cost table run locally in your browser. The tokenizer loads as a cacheable worker. The text is not sent to this site.

When were prices updated?

The date under the cost table is 2026-10-16, when these rates were copied from the providers' pricing pages. Open the cited links to confirm the current rate before you budget a large workload.

AI token counter

Count tokens and estimate the API cost of a prompt for GPT, Claude, and Gemini models.

In Prompt text or a text file, selected models, expected output tokens, requests per day, and a cached-input share./Out Token, character, and word counts, a token boundary preview, and a cost table per model.

Runs locally

Prompt

Nothing is uploaded. The tokenizer runs in a worker.

Drop a .txt, .md, or .json file

Output tokens

Requests / day

Cached input 0%

Tokens

Characters

Words

Model	Request	Month
GPT-6.1 Sol	$0.0050	$15.00
Claude Sonnet 5.5 estimate ≈ o200k × 1.0 ·	$0.0050	$15.00
Gemini 4 Argon estimate ≈ o200k × 1.0 ·	$0.0050	$15.00

Model

Request

Month

GPT-6.1 Sol

$0.0050

$15.00

Claude Sonnet 5.5

estimate ≈ o200k × 1.0 ·

$0.0050

$15.00

Gemini 4 Argon

estimate ≈ o200k × 1.0 ·

$0.0050

$15.00

Prices updated 2026-10-16. Confirm the linked provider pages before a large budget.

Markdown

2| --- | ---: | ---: | ---: | ---: |

3| GPT-6.1 Sol | $0 | $0.0050 | $0.0050 | $15.00 |

4| Claude Sonnet 5.5 | $0 | $0.0050 | $0.0050 | $15.00 |

5| Gemini 4 Argon | $0 | $0.0050 | $0.0050 | $15.00 |

How to use it

1Paste your prompt or fileDrop a .txt, .md, or .json file, or paste the prompt. The count updates in the browser.
2Pick models and expected outputChoose the models to compare, set the output tokens you expect back, and the cached-input share.
3Compare cost per request and per monthThe table prices input and output separately. Copy it as markdown when you want to keep the estimate.