AI Token Calculator
Convert words and characters into tokens, then straight into money. Useful for sizing a prompt, checking whether a document fits a context window, or pricing a batch job.
Inputs
About 1.33 for English prose; higher for code and non-Latin scripts.
Result
998
tokens
$0.0025 per send · $2.49 for 1,000 sends
- Tokens per send
- 998
- Approximate characters
- 4,275
- Cost per send
- $0.0025
- Total tokens
- 997,500
- Total cost
- $2.49
- Share of a 128k context window
- 0.8%
Disclaimer: Model availability, pricing and specifications may change. Verify pricing directly with the provider before making business decisions. Preset rates reflect published list prices last reviewed August 2026; choose “Custom pricing” to enter your own rates.
Formula
- Tokens = words × tokens per word
- Cost = (tokens ÷ 1,000,000) × price per 1M tokens
- Total = cost × number of sends
Methodology
Tokenisers split text into sub-word pieces rather than whole words. English prose lands near 1.33 tokens per word (about 4 characters per token); code, JSON, and languages that do not use the Latin alphabet run considerably higher, sometimes two or three tokens per word.
Because the ratio varies by tokeniser, the field is editable. If you have measured your own average with a tokeniser library, enter that number and the estimate becomes close to exact.
The context-window share assumes a 128,000-token window, a common size across current frontier models. Anything above 100% has to be chunked or summarised before it will fit.
Example: pricing a 750-word document
- A 750-word article at 1.33 tokens per word is about 1,000 tokens.
- At $2.50 per million tokens, one send costs $0.0025.
- Sending it 1,000 times — once per user, say — costs about $2.50 in total.
Frequently asked questions
How many words is 1,000 tokens?
Roughly 750 words of ordinary English, or about 4,000 characters. Dense technical text, code and non-English scripts produce more tokens for the same word count.
Do spaces and punctuation count as tokens?
They are folded into tokens rather than counted separately — a leading space is usually part of the following word's token — but they do add to the total, which is why the character estimate runs ahead of a naive word count.
Are input and output tokens counted the same way?
The counting is identical; only the price differs. Output tokens typically cost three to five times more than input tokens on the same model.
Why does my provider report more tokens than this?
Chat APIs add a few tokens of formatting per message, and tool or function definitions are billed too. Expect a small overhead on top of the raw text estimate.
Are these prices always up to date?
The presets reflect widely published list prices and are reviewed periodically, but providers adjust them and offer discounts for batch processing, cached input and committed spend. Model availability, pricing and specifications may change. Verify pricing directly with the provider before making business decisions.
What happens when a provider launches a new model?
Model names and rates live in one central configuration, so a new model appears in every relevant dropdown as soon as it is added. Page URLs stay provider-based rather than version-based, so no link ever goes stale when a model generation changes.
Does the calculator send my data anywhere?
No. Every calculation runs in your browser, so nothing you type is uploaded, stored or logged.
Related AI calculators
AI Cost
OpenAI Cost Calculator
Estimate what running OpenAI models costs before the invoice arrives. Pick a model from the dropdown — the list is kept current and includes legacy tiers — then enter typical input and output token counts and your daily request volume for an instant per-request, per-day and monthly figure.
AI Cost
AI API Cost Calculator
Turn API traffic into a budget. Choose a model from any major provider — OpenAI, Anthropic, Google, xAI, Meta, Mistral, Cohere or Amazon Bedrock — then enter average tokens per call and expected daily calls to see the monthly bill.
AI Cost
Claude Cost Calculator
Price Claude usage across Opus, Sonnet and Haiku. Enter typical prompt and response sizes with your daily volume to see which tier your workload can actually afford.
AI Cost
Gemini Cost Calculator
Work out Gemini API costs across the Pro and Flash tiers. Enter token counts and daily volume to see the monthly total, and compare tiers by switching the model.
AI Cost
LLM Cost Calculator
A provider-neutral calculator for any large language model. Pick a preset from OpenAI, Anthropic, Google, xAI, Meta, Mistral, Cohere or Amazon Bedrock, or enter your own input and output rates, to compare models on identical traffic.
AI Cost
AI Chatbot Cost Calculator
Chatbots get expensive because every turn resends the conversation so far. This calculator models that growing history and returns a realistic cost per conversation and per month.