PRIVATE, IN-BROWSER CALCULATOR
Paste once. Compare model costs clearly.
Total = input cost + output cost
0 selected
LIVE COMPARISON
1 request
| Model | Input / 1M | Cached / 1M | Output / 1M | This request | Total |
|---|
THE SCOPE
This AI token calculator compares standard text-generation rates from OpenAI, Anthropic, Google, DeepSeek, xAI, Perplexity, Mistral, and Cohere. Every listed price includes a direct official source and the date it was checked.
THE MATH
cost = input_tokens × input_rate ÷ 1,000,000
+ output_tokens × output_rate ÷ 1,000,000
Volume totals multiply the per-request result by your request count. “Per day” also shows an estimated 30-day month; it does not predict traffic.
TOKEN COUNTING
The live text count uses the documented English-language planning rule of approximately four characters per token. Exact tokenization differs by model and language, so the input field stays editable for a provider-tokenizer result.
PRIVACY
Counting and price calculations run entirely in your browser. There is no account, server request, history, batch upload, analytics payload, or file export.
PRICE PROVENANCE
Rates are embedded in app.js; the page makes no live pricing request. Recheck the linked official pages before committing a production budget.
QUICK ANSWERS
Both phrases describe the same intent here: count or enter tokens, then compare the resulting API cost across selected AI models.
The pasted-text figure is an estimate for planning. For an exact billed count, paste the token total returned by the provider’s tokenizer or API into the editable input-token field.
Some providers discount repeated prompt content served from a prompt cache. The table shows the published cached-input rate when the provider lists one; the request total uses the standard input rate.
No. The calculator has no backend and does not save history. Reloading the page clears the current calculation.