Home/Token counter
Token Counter
Paste any text to count its tokens, characters and words, and see what sending it costs on 10 popular models, from $0.10 to $10.00 per million input tokens. Counting runs in your browser; your text is never uploaded. Prices verified October 2026.
What This Text Costs as Input
Sending the text once as a prompt, at each model’s standard input rate. The answer is billed separately at the output rate.
| Model | Input /1M | This text | × 1,000 |
|---|---|---|---|
| GPT-6 LunaOpenAI | $0.10 | — | — |
| DeepSeek V4.1 FlashDeepSeek | $0.30 | — | — |
| Gemini 3.8 FlashGoogle | $0.75 | — | — |
| Claude Haiku 4.5Anthropic | $1.00 | — | — |
| GPT-6.1 SolOpenAI | $2.00 | — | — |
| Claude Sonnet 5.5*Anthropic | $2.00 | — | — |
| Gemini 3.1 ProGoogle | $2.00 | — | — |
| Grok 4.7xAI | $2.00 | — | — |
| Claude Opus 5.5*Anthropic | $4.00 | — | — |
| GPT-6 AstraOpenAI | $10.00 | — | — |
Counts use OpenAI’s o200k_base tokenizer. Other providers count the same text differently. * Anthropic says Claude models from 4.7 on produce about 30% more tokens for the same text than earlier Claude models (Anthropic pricing docs), so expect a higher count there. Google and the other providers publish no conversion factor.
Pricing a whole request, answer included? Use the prompt cost calculator, or compare every model per 1M tokens.
Token Counter — FAQ
Paste or type it into the box above. The count updates as you type, using the o200k_base tokenizer from OpenAI, and you also get the character and word count. Counting runs entirely in your browser: the text is never uploaded.
OpenAI's rule of thumb for English is that one token is about 4 characters, or about three quarters of a word, so 100 tokens are roughly 75 words. Code, numbers and many non-English languages usually take more tokens per word, so paste a sample of your own text for a real count.
No. Every provider uses its own tokenizer, so the same text gives a different count on each. Anthropic says Claude models from 4.7 on use a tokenizer that produces about 30% more tokens for the same text than earlier Claude models. Google and the other providers publish no conversion factor, so treat the count here as an approximation for non-OpenAI models; for an exact billed count, use the provider's own token-counting API.
On current models, 1,000 input tokens cost $0.000100 to $0.0100 and 1,000 output tokens $0.000500 to $0.0500 (prices verified October 2026). Among the popular models above, input runs from $0.10 per million on GPT-6 Luna to $10.00 on GPT-6 Astra.
The text you paste is what you send to the model, so it is billed at the input rate. The model's answer is billed separately at the output rate, which is 2× to 8.3× the input rate on current models. To price a full request with its answer, use the prompt cost calculator.