Prompt Engineering

Tokens vs Words — How Many Tokens per Word

Word count feels intuitive, but models bill tokens. This shows tokens per word on a real article so you can convert between the two with eyes open.

Open in Token Counter

Overview

English runs roughly three tokens for every four words, but "roughly" carries a lot of weight: punctuation, rare words, and formatting all push the ratio around. This loads an article-length text and reports the tokens-per-word figure directly, so a word count you already have becomes a token estimate you can trust within a range. Words are a writer's unit; tokens are the model's — this is the bridge between them.

How to use this resource

Count a real text

Words and tokens reported together on an article.
Read tokens per word

English sits near 4 words to 3 tokens — with drift.
Convert your own counts

Apply the ratio to a word count you already have.

Why This Works

A direct tokens-per-word figure turns word counts into token estimates
Real text shows the ratio drifting with punctuation and rare words
The range respects that the conversion is approximate, not exact

Best for

Converting a known word count to tokens
Estimating from a word-count limit
Understanding the words-to-tokens ratio

Not for

Counting words for an essay or post limit — use the Character Counter
Context-window fit decisions — use the Context Window Estimator

FAQ

How many tokens is a word in English, and can I trust that ratio for billing?

English averages roughly "~0.75 words per token," and this report prints a tokens-per-word figure directly on real prose. Treat it as a range, not a billing number: the ESTIMATION NOTES stress these are "character-based ESTIMATES, not tokenizer output" that vary across GPT, Claude, and Gemini. Convert a word count you already have into a token band, then verify against the provider.

Should I use the token count or the fit check to decide if my prompt fits a model?

Use the fit check for that. This report's MODEL NOTES call the per-model figures "a count, not a fit check — for 'will it fit?' use the Context Window Estimator." The Token Counter answers how many tokens and what they'd cost — the per-1,000-calls line ($13.42-$13.74 here) is where cost gets real — while context-window headroom is a separate tool's job.

Customize This Resource

Opens this text in Token Counter. Count to get the full token and cost report — then adjust the model and assumed response length.

Open in Token Counter

Prompt Template

Copy it as-is, or use Open in Token Counter to load it pre-filled and customize it with your own context.

TOKEN COUNT REPORT

TOKEN ESTIMATE
- Estimated tokens: ~528 (range 475–581)
- Characters: 1,990
- Words: 348
- Detected content type: Prose
- Tokens per word (approx): 1.5
- Tokens per character (approx): 0.3

COST ESTIMATE — Claude Sonnet
- Pricing (approximate, June 2026): input $3.00/1M tokens · output $15.00/1M tokens
- Input cost (this prompt): $0.001425–$0.001743
- Assumed response: Medium response (~800 tokens) -> output cost $0.0120
- Combined per call: $0.0134–$0.0137
- Per 1,000 calls: $13.42–$13.74

MODEL NOTES
- Anthropic tokenizer (same family behavior as Opus) at a lower price point.
- Same text, estimated per model: GPT-5 ~503 · Claude Opus ~528 · Claude Sonnet ~528 · Gemini Pro ~503 (a count, not a fit check — for "will it fit?" use the Context Window Estimator).

USAGE GUIDANCE
- For scale on Claude Sonnet: a short prompt ≈ 75, a medium prompt ≈ 531, a large prompt ≈ 3,182 tokens.
- This text is closest to a medium prompt.
- A single call looks cheap; the per-1,000-calls line is where token cost becomes real — budget on volume, not on one request.

ESTIMATION NOTES
- A token is not a character and not a word — it is a sub-word chunk. English averages ~4 characters / ~0.75 words per token.
- Estimates vary by tokenizer: the same text tokenizes differently on GPT, Claude, and Gemini — that is why this is a range, not a single number.
- Language matters: CJK and many non-Latin scripts use more tokens per character than English.
- Code differs from prose: symbols, indentation, and punctuation push code to more tokens per character.
- These are character-based ESTIMATES, not tokenizer output. Pricing is approximate as of June 2026; providers change rates — verify before relying on a number.

More resources from Token Counter

Resource

Estimate Token Usage Before You Run It

Know how many tokens a job will consume before you send it — input plus an assumed response, costed per call and at scale.

Prompt Engineering

Resource

Token Counter for AI Prompts

Paste a prompt, get an honest token estimate — a range, not a fake-precise number — plus the cost across GPT, Claude, and Gemini.

Prompt Engineering

Resource

Calculate AI API Cost for a Prompt

Turn a prompt into a dollar figure: input cost, output cost, combined per call, and the number that actually matters — cost per 1,000 calls.

Prompt Engineering

Resources that pair well

Resource

Estimate Token Budget — Plan Before You Paste

Token budget planning for real workloads: how much of the window a transcript actually consumes, what is left for the answer, and how much headroom remains.

Prompt Engineering

Resource

Message Too Long — the Fix That Doesn't Butcher Content

The "message too long" error has a structural fix: split at paragraph boundaries into sequenced chunks with wait rules, instead of pasting fragments and hoping.

Prompt Engineering

Resource

Prompt Cleanup Examples (Before & After)

A set of before-and-after examples showing exactly what prompt cleanup removes — and what it deliberately leaves alone.

Prompt Engineering

Related tools

Tool

Token Counter

Estimate how many tokens a prompt is and what it costs — honest ranges across GPT, Claude, and Gemini, with per-call and per-1,000-call pricing.

Prompt Utilities

Tip: Save time by exploring related resources and tools that integrate with this resource.