Free Claude token counter · Runs in your browser

Count Claude tokens — and what they cost

Paste any text or code and instantly see the token estimate plus the price on Opus, Sonnet, and Haiku — as input and as output, including cache rates. No signup, nothing uploaded.

text

Paste something to count. Rough rule: 1 token ≈ 4 characters of English, less for code.

100% client-side · nothing uploaded · calibrated for prose and code

Claude API pricing

The full price table, one place

Per million tokens, July 2026. Output always costs ~5× input; cache reads cost 90% less than fresh input — the two facts that explain most Claude bills.

🧠

Claude Opus

Input $15 / MTok · Output $75 / MTok
Cache write $18.75 · Cache read $1.5
The heavyweight — for the hardest reasoning and agentic work.

⚖️

Claude Sonnet

Input $3 / MTok · Output $15 / MTok
Cache write $3.75 · Cache read $0.3
The default — best cost/intelligence balance for coding.

Claude Haiku

Input $1 / MTok · Output $5 / MTok
Cache write $1.25 · Cache read $0.1
The economy tier — classification, extraction, high-volume calls.

How tokens work

Four facts that make bills predictable

~4 characters per token

English prose averages 3.5–4 characters (≈0.75 words) per token. 'The quick brown fox' is 4 tokens; this whole sentence is about 12.

Code tokenizes heavier

Brackets, operators, and camelCase identifiers split into more tokens per character — budget 25–40% more than prose. This counter detects code and adjusts.

You pay both directions

Every call bills the entire prompt (input) plus the response (output) — and in a conversation, the whole history is re-sent as input each turn. Long chats compound.

Caching changes everything

Repeated context (system prompts, codebases) can be cached: 25% extra to write, 90% off to read. It's why Claude Code can hold your repo in context affordably.

FAQ

Token counting questions

How do I get an exact token count?+
Use the API's free count_tokens endpoint — it runs the real tokenizer server-side. For planning, budgeting, and prompt-size sanity checks, this estimator's ±5% is plenty.
Do Claude and GPT count tokens the same way?+
No — every model family has its own tokenizer, so the same text produces different counts. The differences are usually within ~10–20% for English, but never assume counts transfer between providers.
How many tokens fit in Claude's context window?+
Claude models offer 200K-token context windows (with 1M-token tiers available on some models). 200K tokens is roughly 150,000 words — about two novels — but remember: everything in context is billed as input on every call.
Why did my agent workflow cost more than this estimate?+
Agents loop: each tool call re-sends the conversation as input, and system prompts, tool definitions, and file contents all count. Multiply per-call cost by turns, and use cache pricing for the repeated prefix — or just measure real usage with ccusage and our Claude Code Wrapped tool.
Is my text stored or sent anywhere?+
No. The counter is client-side JavaScript — your text never leaves the browser.
Go further

Spend those tokens on real work

The ClaudeThings kits are token-efficient by design — tight skills, scoped agents, and commands that get more done per call, for engineering and marketing.

🛠️
Most popular

Engineering Kit

Battle-tested Claude skills & workflows for builders.

  • 58 agents · 61 skills · 159 commands
  • Security, testing & code-review skills
  • Every skill formatted to spec
  • Lifetime updates
Get the Engineering Kit
📣
For founders

Marketing Kit

Turn Claude into your growth team.

  • 31 agents · 42 skills · 32 commands
  • Content, SEO & email-sequence skills
  • Landing-page & funnel templates
  • Lifetime updates
Get the Marketing Kit

Or grab all 89 agents, 103 skills & 181 commands in the Complete Bundle →