Token counterCalculators

Token counter. Size a prompt before you send it.

Paste any text. Characters, words and lines are exact. The token count is an estimate from a rule of thumb, not a model's own tokenizer, shown with a low and high bound. Pick a model to see the context share and cost.

Cost to you

Free

Runs in your browser. No signup, no API call, no limit per hour.

browser
runs in: none
API calls: none
account

Use it now

Use the Token counter here

It runs in your browser. Nothing you type is sent anywhere, there is no signup and no limit.

Text or prompt

Paste a prompt, a document or a chat log. Counts update as you type.

Model

$2.00 input and $10.00 output per million tokens. Context 1,000,000 tokens.

Counts

Tokens (estimate): ~80
Characters: 311
Words: 60
Lines: 3

Estimate range 78 to 82 tokens. The real count depends on the model's tokenizer.

under 0.01% of the 1,000,000 token context window of Anthropic: Claude Sonnet 5.5.

Cost if sent as input: $0.00016
Cost if written as output: $0.0008

Characters, words and lines are exact. Tokens are an estimate from character and word counts, not the model's own tokenizer, so check with the provider's token counting endpoint when a bill depends on it. Prices read from the OpenRouter model list through looot on 2026-10-13. Runs in your browser, nothing is sent anywhere.

Run it from your agent

Copy one prompt, give it to your agent.

Fill in the placeholder, paste it, and the agent runs the same job with current data on your looot account and shows the cost.

You give it: pasted text and a model.
Billed per call, nothing while your agent is idle.
The agent shows the price before it calls.

Prompt: Token counter

Use looot to size this prompt for [model]: [paste the prompt]. 1. Set up https://looot.ai/skill.md. 2. Run the OpenRouter model list endpoint to read the model's context length and price per million tokens. 3. Count the prompt's tokens with the model's own tokenizer if you have it, or estimate. 4. Return tokens, share of the context window and the cost per call.

Give this to your agent

Setup command

set up looot - https://looot.ai/skill.md

What comes back

What your agent gets from token counter

Field names as the providers return them. The exact set varies by endpoint, and your agent reads it from the catalog.

estimated tokens with a low and high bound
characters, words and lines
share of the context window
cost as input and as output

Use it for

What people do with token counter

Fit the context
Check a long document fits the model before the call fails.
Price a prompt
See what a system prompt costs on every call at your model's price.
Trim the fat
Cut words and watch the estimate drop as you type.
Plan agent memory
Size the history an agent re-sends on each step.

Set up once

Three steps. Your agent does the rest.

Paste one prompt into your agent

Set up looot

Done, looot is connected.

Copy one prompt from /connect into Claude Code, Cursor, or any MCP agent. No SDK, no dashboard click-through first.

Your agent finds the right operation

Hunter$0.0265

Exa$0.0090

Diffbot$0.0032

The agent finds the operation it needs and sees the price first. It calls nothing blind.

You watch what it cost

Hunter$0.0265

Exa$0.0090

Diffbot$0.0032

Total$0.0387

Every call settles for its quoted price and lands in the ledger. You always know what ran and what it cost.

Keep going

Calculators
LLM token cost calculator
Cost per call and per month for any model in the OpenRouter list, from your own token counts.
Free, no signupno limit
Research
AI model comparison
A sortable table of AI models by input price, output price and context size, from a dated model list.
Free, no signupno limit
Calculators
AI agent cost estimator
Cost per task and per month for an agent that loops, counting the history it re-sends on every step.
Free, no signupno limit

FAQ

Questions about token counter

Paste it in the box. The page shows an estimate with a low and high bound. For an exact count, use the model provider's own tokenizer or token counting endpoint.

Each model family has its own tokenizer, and this page bundles none. It estimates from characters and words, which is close for English prose and rougher for code and other languages.

In English prose a word is often a little more than one token. Code, numbers, rare words and many other languages take more.

Yes. Each of those characters counts as one token in the estimate. Real counts vary by tokenizer.

From the OpenRouter model list, read through looot on the date shown under the counter.

No. Counting runs in your browser. Nothing is uploaded or stored.

Yes. No account and no limit.

For agents

tool: Token counter
category: Calculators
input: Pasted text and a model
output: estimated tokens with a low and high bound, characters, words and lines, share of the context window, cost as input and as output
price: per call, read live from the catalog on the web page
web tool: runs in the visitor's browser, no API call, free and unlimited
schedule: on demand
setup: https://looot.ai/skill.md
llms.txt: https://looot.ai/llms.txt
updated: 2026-10-14

markdown /free-tools/token-counter.md

Run token counter from your own agent.

Start your workspace, top up, and paste the prompt into the agent you already use. It runs the job with live data and shows the cost first.

Start your workspace See all free tools