Count and price it.
Paste-to-Token & Prompt Cost Counter
A standard English word equals roughly 0.75 tokens. Paste your text below to instantly calculate the exact token count and API input cost for leading LLMs like GPT-4o, Claude 3.5, and Gemini 1.5 without sending your data to any external server.
Character count, word count and the price of this exact prompt appear the moment you paste — no calculate button, no upload, no account.
Private · Local · No account
Your prompt in, its token count and input cost out.
Every model bills per million tokens, so the first question is always how big the prompt is. This counter estimates tokens from your words at the standard 0.75 ratio, then prices that total against the published input rates for GPT-4o, Claude 3.5 Sonnet and Gemini 1.5 Pro — instantly, as you type.
Count my tokens- Characters, words and tokens live
- GPT-4o, Claude 3.5 & Gemini 1.5 pricing
- Text never leaves the browser
Paste, count, price
100% Private: Processed locally in your browser.| Model | Provider | Input $ / 1M tokens | Cost for this text |
|---|---|---|---|
| GPT-4o | OpenAI | $5.00 | $0.0000 |
| Claude 3.5 Sonnet | Anthropic | $3.00 | $0.0000 |
| Gemini 1.5 Pro | $3.50 | $0.0000 |
Cost = (estimated tokens ÷ 1,000,000) × input price per 1M tokens. These are published input list prices — cached-input, batch and long-context discounts are not applied, so treat the figure as a ceiling for this prompt alone.
How the token estimate works
There is no universal token. Each provider splits text with its own tokenizer, so any count derived from a word total is an estimate rather than a bill. This counter uses the standard English ratio — 1 token ≈ 0.75 words, or about 4 characters — and rounds up so a partial token is never dropped.
Character count includes spaces and line breaks and counts an emoji as one character. Word count splits on whitespace runs, so text separated only by a line break counts as two words, and leading or trailing spaces are ignored.
The cost table prices the input side only: GPT-4o at $5.00, Claude 3.5 Sonnet at $3.00 and Gemini 1.5 Pro at $3.50 per 1 million input tokens. These are published list rates — cached-input, batch and long-context discounts are not applied, and output tokens are not counted, because this tool prices the text you paste, not the answer a model generates. Confirm current rates with the provider before you budget.
The result is an estimate for planning, not an invoice or financial advice.
The formulas
Estimated tokens = Math.ceil(words ÷ 0.75)
Cost = (tokens ÷ 1,000,000) × input price per 1M tokens
Words = whitespace-separated runs in the pasted text
Worked example: a 300-word prompt is Math.ceil(300 ÷ 0.75) = 400 estimated tokens. On GPT-4o at $5.00 per million input tokens that is (400 ÷ 1,000,000) × $5.00 = $0.0020; on Claude 3.5 Sonnet at $3.00 it is $0.0012, and on Gemini 1.5 Pro at $3.50 it is $0.0014.
FAQ
Frequently asked questions
How many words is one token?
On average, 1 token is approximately 4 characters or 0.75 words in English.
Is this token counter private?
Yes, all token and cost calculations are performed locally in your web browser. No text is uploaded to any server.