Document Token & Cost Estimator
Turn a document's text, word count or page count into an estimated token total and the cost to send it as input.
Before you feed a report, contract or slide deck to a model, it helps to know roughly how many tokens it holds and what that costs. This tool answers "how many tokens in a PDF" without any upload: paste the extracted text, or work from a word count or a page count, and it estimates the token total and the input cost against the model you pick. It doubles as a pdf to tokens calculator and a word count to tokens converter — pick a document type so the words-per-page and tokens-per-word match prose, legal text, slides or code. The count is a heuristic, not an exact tokenizer, so read it as a close planning figure rather than the number your bill will show.
Read the guide: How Many Tokens Are in a PDF, Doc or Slide Deck?Your document
Estimated tokens
0
≈ $0 to send as input on GPT-4o
- Words
- 0
- Input cost
- $0
- Characters
- 0
Token counts are an estimate, not an exact tokenizer — different models split text differently, so expect ±10–20%. Prices updated January 2026; input-only, output is billed separately. For a PDF or Word file, first extract its text (copy it out or export to .txt) and count that — this tool does not read binary document formats. Verify rates on the provider's pricing page before you budget.
How it works
- 1
Paste text or switch to a count
In Paste mode, drop the document's text (or a .txt file) and the word, character and token counts update live. In By page count mode, enter a word count or a page count instead.
- 2
Match the document type
When counting by pages, pick prose, legal, slides or code. Each sets a realistic words-per-page and tokens-per-word, since a dense contract and a sparse slide deck size very differently.
- 3
Price it against a model
Choose the model you will send it to. The estimate multiplies the token count by that model's input rate, so you see both the token total and the dollar cost to send the document once.
Instant & 100% private — nothing is uploaded
Every calculation runs locally in your browser. The prompts, token counts and numbers you enter stay on your own device and are never sent to a server — nothing is stored, logged or shared.
Frequently asked questions
- How many tokens are in a PDF?
- It depends on the text inside, not the file. A typical prose page is around 500 words, which is roughly 650 tokens, so a 10-page PDF lands near 6,500 tokens. Extract the PDF's text and paste it, or use the page count with a matching document type, to get a closer figure. Dense legal text runs higher per page; slides run much lower.
- Why can't I upload the PDF itself?
- This tool reads plain text, not binary PDF or Word files, and it does not add any parsing library. To count a PDF, open it and copy the text out, or export it to a .txt file and drop that in. Counting the extracted text is what actually determines the token total anyway.
- How is the token count estimated?
- In paste mode it blends two rules of thumb — about four characters per token and 1.33 tokens per word — into one figure. In count mode it multiplies words by a tokens-per-word rate chosen for the document type. Neither is the model's exact tokenizer, so treat the result as an estimate within about 10–20% for ordinary English.
- Is this exact, and is my document private?
- It is an estimate, not an exact count — every model uses its own vocabulary, and code, tables and non-English text split differently. Everything runs in your browser: pasted text and dropped files are read locally and never uploaded. Prices come from a dated table on the page, so verify current rates on the provider's pricing page before you budget.
Important
For planning and estimates only. Prices come from a published rate table dated on the page; providers change pricing without notice, and token counts here are approximations. Confirm against the provider’s own pricing before you budget or commit.
More tools
More from the Hivly network
Free sister tools on our other sites.
