How anonymizer compares
An honest look at how anonymizer stacks up against other ways to strip PII from documents.
Feature comparison
| Tool | Price | Offline | Formats | RU / EN | Output | Open-source |
|---|---|---|---|---|---|---|
| anonymizer | Free (AGPL) | Yes | .docx / .pdf / .xlsx | RU · EN · ES* | Structured tokens | Yes |
| CamoText | Paid ($59–99) | Yes | .docx / .pdf / .txt / .csv / .xlsx | EN + EU | — | No |
| anonym.plus | Paid (€0–499) | Yes | .docx / .pdf / .xlsx / .csv / images | Multi | — | No |
| RedactChat | Free tier | Browser (text only) | Text / prompts | EN | — | No |
| DOCXдодыр | Free | Yes | .docx only | RU | — | No |
| Microsoft Presidio | Free (OSS) | Yes | Text API (SDK) | EN | Tokens | Yes (dev-level) |
| Manual Find & Replace | Free | Yes | Any | Any | Manual | n/a |
* Spanish is an optional add-on (downloadable model); Russian and English are built-in.
Why anonymizer
anonymizer is the only option that combines all five: free & open-source (AGPL), fully offline, native .docx / .pdf / .xlsx support, Russian and English entity detection out of the box (plus an optional Spanish pack) — ИНН, СНИЛС, passport numbers, Cyrillic names alongside EN/EU PII — and reversible structured tokens — so the original text can be reconstructed from the session token map without any manual find-and-replace in reverse.
Why Find & Replace isn't enough
Word's Find & Replace only catches what you explicitly type. It misses spelling variants and case differences, entities buried in headers, footers, comments, and revision history, document metadata (author, company, last-saved path), and any name or ID you simply forgot to list. The result is inconsistent token naming and residual PII that looks clean on screen but survives a metadata strip.
FAQ
Is there a free offline alternative to CamoText or anonym.plus?
Does anonymizer support Russian documents?
Get started
uv tool install docs-anonymizer Full documentation — CLI reference, format coverage, detector configuration — at anonymizer.site/docs.