Your Company Data Could Be Worth $100K to $2M
AI companies are paying established businesses for real internal data: SOPs, knowledge bases, CRM records, project histories, QA processes and team communications. The documentation you already have can become a new revenue stream. If you want to sell, answer three quick questions to see if your company qualifies.
- Open to every industry
- US companies first, then other Western markets
- No new work, you license what already exists
Check if your company qualifies
Good news. Your company meets the requirements for the Enterprise Data Partnership. Apply below to find out what your data is worth.
Apply for the Data Partnership →Bulk Document Token Counter — Count Words and AI Tokens in Many Files
Most token counters only take one block of pasted text. That does not help when you have a folder full of SOPs, reports and proposals. This tool counts them all at once. Drop in as many files as you like, or choose a whole folder. It reads Word, PowerPoint and Excel files, plus text, Markdown, CSV, JSON and HTML. For each file you see the words, characters and estimated AI tokens, and the totals for the full set appear at the top. You can sort the table and copy the results as CSV. Files are read inside your browser, so nothing is uploaded and private documents stay private. Use it to estimate AI processing costs, check if a document set fits a model's context window or measure the size of your company knowledge base.
or click to choose files (.docx, .pptx, .xlsx, .txt, .md, .csv, .json, .html)
| File | Words | Characters | Est. tokens |
|---|
Frequently Asked Questions
How do I count tokens in many documents at once?
Drop all your files into the box above or choose a whole folder. The tool reads each file, pulls out the text and counts words, characters and estimated tokens. You get a table per file plus a total for the whole set.
Which file types are supported?
Word (.docx), PowerPoint (.pptx), Excel (.xlsx), plain text, Markdown, CSV, JSON, HTML, XML and log files. Older .doc, .ppt and .xls files and PDFs are not supported. Save them in the newer format first.
How accurate is the token count?
The tool uses the common rule of about 4 characters per token for English text. Exact counts vary by model, often by 5 to 10 percent. For sizing a dataset or estimating AI costs, this is close enough.
Are my files uploaded?
No. Every file is opened and counted inside your browser. Nothing is sent to a server, so it is safe to use with private company documents.
Why count tokens across a whole document library?
Token totals tell you what it will cost to process documents with an AI model, whether they fit in a context window and how large your knowledge base really is. Data buyers also measure company archives in tokens.
What Is a Token?
AI models do not read whole words. They split text into tokens, which are often parts of words. In English, one token is about 4 characters or three quarters of a word. AI pricing, context limits and data deals are all measured in tokens.
Context Window Check
Many current AI models accept 128,000 to 1,000,000 tokens in one request. If your document set is larger, you will need a search based setup that sends only the relevant parts. The total above shows which side of the line you are on.
Estimating AI Costs
Multiply the token total by your model's price per million input tokens. For example, 20 million tokens at $3 per million costs about $60 to process once. Repeated processing, summaries and outputs add to that, so plan with a buffer.
Sizing Your Knowledge Base
Run the tool on your shared drive or SOP folder to see how much written knowledge your company holds. Large, well organized document sets in English are valuable, both for internal AI tools and for licensing to AI companies.
More Free Tools
Random Weight Generator
Generate random probability weights for any number of items using Dirichlet distribution sampling.
DPMO Calculator
Calculate defects per million opportunities, yield and Six Sigma level, or convert between them.
Computer Virus Name Generator
Generate realistic fake malware names in antivirus naming conventions with threat profiles and severity ratings.
Rotating Shift Schedule Generator
Generate Panama, Pitman, 4-on-4-off, or continental shift schedules with CSV export and print support.