allocateLLM token counter
Paste a prompt or document to count its tokens. See what it costs as input on every model, and whether it fits each context window.
Prices checked 7 Jul 2026, against published provider rates.
How it works
Common questions
How accurate is the token count?
It blends character and word heuristics tuned to modern tokenizers and typically lands within 10 to 15% of exact counts for English prose; code and non-Latin scripts vary more. For billing-exact numbers, use each provider's tokenizer; for planning and budgeting, this is enough.
How many tokens is a page of text?
A typical single-spaced page of around 500 words is roughly 650 to 700 tokens. A thousand tokens is about 750 words of English prose.
Why do token counts differ between models?
Each model family trains its own tokenizer with a different vocabulary, so the same text splits differently: usually within 10 to 20% of each other for English, more for other languages and code.
Do I pay for input and output tokens differently?
Yes. Output tokens typically cost 4 to 5 times more than input because generation is sequential. This counter prices your text as input; use the price comparison tool for full workload costs.
Is my text uploaded anywhere?
No. The counter runs entirely in your browser, which matters when you're counting a document full of customer data.
More free tools
Allocate meters every token per route and per agent, with hard caps, so what you counted here is what you spend in production.