LLM token counter and context checker
Paste text to estimate how many tokens it uses, which Claude, GPT, and Gemini models it fits in, and what one request would cost.
Context windows checked against provider documentation on October 21, 2026, prices on October 20, 2026. Estimates are approximate. Your text never leaves your browser.
Your text
Estimated tokens
- Estimated input tokens
- 0 Paste some text to see an estimate.
- Characters
- 0
- Words
- 0
Which models it fits
Paste some text to see the best pick for each vendor. Costs use the middle estimate for one request, at standard prices.
Anthropic Claude
- Claude Haiku 5.5$0.00051M context · uses 0.1% of the window
- Claude Haiku 4.5$0.0050200K context · uses 0.5% of the window
- Claude Sonnet 5.5$0.01001M context · uses 0.1% of the window
- Claude Opus 5.5$0.02001M context · uses 0.1% of the window
- Claude Fable 5.1$0.05001M context · uses 0.1% of the window
OpenAI GPT
- GPT-6 Luna$0.00051.05M context · uses <0.1% of the window
- GPT-6.1 Sol$0.01001.05M context · uses <0.1% of the window
- GPT-5.5$0.03001.05M context · uses <0.1% of the window
- GPT-6 Astra$0.05001.05M context · uses <0.1% of the window
Google Gemini
- Gemini 3.5 Flash-Lite$0.00251.05M context · uses <0.1% of the window
- Gemini 3.8 Flash$0.00371.05M context · uses <0.1% of the window
- Gemini 4 Argon$0.01001M context · uses 0.1% of the window
- Gemini 3.1 Pro (preview)$0.01201.05M context · uses <0.1% of the window
How many tokens is it?
Rough conversions for English text, using one token per three quarters of a word. Paste your own text above for a closer estimate.
| Length | Words | Approx. tokens |
|---|---|---|
| A long email | 300 | 400 |
| One page of text | 500 | 667 |
| A 10-page document | 5,000 | 6,667 |
| A 100-page report | 50,000 | 66,667 |
| A typical novel | 90,000 | 120,000 |
Tokens, explained
- What is a token?
- A token is the unit a language model reads and writes, usually a short word or a piece of a longer one. A common short word like “the” is usually one token, while a long or unusual word is split into several. Providers bill by the million tokens and limit each request by tokens, not by words or characters.
- How many tokens is 1,000 words?
- About 1,300 for ordinary English text, using the rule of thumb that one token is roughly three quarters of a word. Technical writing, code, and text in other languages usually take more.
- Why is the count an estimate?
- Each provider uses its own tokenizer, so the same text can produce different counts on Claude, GPT, and Gemini. This tool estimates from character types and shows a range rather than a single exact number. For exact counts, use the provider’s own counting tools, such as Anthropic’s token counting API.
- Do code and other languages use more tokens?
- Usually yes. Code, JSON, and logs split into more tokens per character than regular text, and many non-English scripts use more tokens per character than English. Choose “Code, JSON, or logs” above for technical text.
- What is a context window?
- The context window is the most a model can handle in one request: your input, the conversation so far, and the response it writes all count toward it. If the input alone is too large, the request fails, so check fit before you build around a long document.
- Do input and output tokens cost the same?
- No. Output tokens are billed at a higher rate than input tokens, often several times higher, which is why the checker asks for an expected response length. The cost calculator can turn token counts into a monthly budget.
I’m Courtney Stephenson, an AI strategy and implementation leader, and I built this tool. Get in touch if you’re planning around context limits and costs.
Want a second opinion on your results?
Book a free 15-minute call and we’ll talk through what this means for your team.