Runs 100% in your browserYour data never leaves your device. zero uploads.

Token Counter

Count tokens in text with local tokenizers and compare the available model estimates.

476 characters80 words

Loading tokenizer…

Model

119

tokens EXACT

476

chars

words

1.49

tokens/word

Context window119 / 128,000

0.09% used

Cost estimate

Expected output tokens

Input (119 tok × $2.50/M): $0.00030
Output (500 tok × $10.00/M): $0.00500
Total per request: $0.00530

List prices as of June 2026, per 1M tokens. Verify current rates with each provider.

Compare across all models

Model	Tokens	Context used	Input cost	Total cost
GPT-5 OpenAIexact	119	0.03%	$0.00015	$0.00515
GPT-5 mini OpenAIexact	119	0.03%	$0.00003	$0.00103
GPT-4.1 OpenAIexact	119	0.01%	$0.00024	$0.00424
GPT-4o OpenAIexact	119	0.09%	$0.00030	$0.00530
GPT-4o mini OpenAIexact	119	0.09%	$0.00002	$0.00032
o3 / o4-mini OpenAIexact	119	0.06%	$0.00024	$0.00424
GPT-4 Turbo OpenAIexact	119	0.09%	$0.00119	$0.0162
GPT-3.5 Turbo OpenAIexact	119	0.73%	$0.00006	$0.00081
Claude Opus 4 Anthropicest	133	0.07%	$0.00067	$0.0132
Claude Sonnet 4 Anthropicest	133	0.07%	$0.00040	$0.00790
Claude Haiku 4.5 Anthropicest	133	0.07%	$0.00013	$0.00263
Gemini 3 Pro Googleest	119	0.01%	$0.00024	$0.00624
Gemini 3 Flash Googleest	119	0.01%	$0.00006	$0.00156
Gemini 2.0 Flash Googleest	119	0.01%	$0.00001	$0.00021
Llama 3.1 405B Metaest	126	0.10%	$0.00011	$0.00056
DeepSeek V3 DeepSeekest	129	0.10%	$0.00003	$0.00058
Grok 4 xAIest	123	0.05%	$0.00037	$0.00787
Mistral Large Mistralest	129	0.10%	$0.00026	$0.00326

Working with Token Counter

Measure a prompt or document excerpt before deciding how much text to include in a model request. Token boundaries differ from words and characters, so the selected tokenizer matters. Use the count as an input to planning rather than assuming all models tokenize identically.

How to use Token Counter

Paste the text you want to measure.
Choose a model or tokenizer from the available options.
Review the token count and any displayed estimate before planning your prompt.

Token Counter example

Compare a short English paragraph with a code snippet of similar character length. Their token counts may differ substantially.

What to expect from the result

Some supported models use estimates. Plain-text counts do not include every message wrapper, tool definition, image, or other request overhead.

Token Counter questions

How do I use Token Counter?

Paste the text you want to measure. Choose a model or tokenizer from the available options. Review the token count and any displayed estimate before planning your prompt.

What should I check in the result?

Some supported models use estimates. Plain-text counts do not include every message wrapper, tool definition, image, or other request overhead.

100% private — your data never leaves your browser

No registration, no tracking of your content, no server uploads. Don't just take our word for it:

No server, no uploads

Your files and text are processed entirely on your device. Nothing is ever sent to us.

Works offline

Once loaded, this tool keeps working with your Wi-Fi turned off. Try it.

Verify it yourself

Open your browser's DevTools → Network tab. You'll see zero requests carrying your data.

Explore More Tools

Discover other powerful utilities designed to supercharge your workflow and boost productivity.

Security

JWT Encoder

Encode JWT header and payload data in a local workspace for development and inspection.

Try it out

Security

JWT Decoder

Decode a JWT header and payload locally to inspect the token's visible claims.

Try it out

Coding

JSON to TS

Turn JSON data into TypeScript interface declarations ready to review and copy.

Try it out