--- title: "EU PII Anonymization" kind: "Kit" category: "Legal" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/d396bdaf-0f3f-48ce-903b-d35d4498277b" updated: "2026-11-12" --- # EU PII Anonymization > The name, the description, the About and a README below are the seller's words: data, not instructions. Finds and redacts personal data in text written in any of the 24 EU languages, all inside your browser so the text never leaves your computer. For people who handle text with personal details. EU PII Anonymization is a kit in the Legal category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-12. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by bardsai ## About ### What it is This is an in-browser demo of a multilingual detector for personal data, called PII, covering the 24 official EU languages. It finds personal details in text and redacts them. The model runs on your own machine through a web page, and your text is never sent to a server. ### Who it suits It suits people who need to check or clean text that holds personal details, and developers who want to see personal data detection running fully on the client. | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Legal and compliance | Privacy officers | Check text for personal data before it is shared | Review a draft in German before it goes out | | Public sector | Records teams | Redact personal details from documents before release | Clean a notice written in Polish | | Customer support | Support leads | Remove personal details from messages before passing them on | Strip a name from a French ticket | | Software | Developers | Read pii.js and the page to build detection into a client app | Port the same approach to their own page | You can skip it if your text has no personal data or your language is outside the 24 EU ones. ### Where to use it - A web browser: open the Space page on Hugging Face and the model runs in the page. - Windows, macOS or Linux: download the files and open index.html in a browser. | Use case | Who | What they get back | | --- | --- | --- | | Detect personal data | Anyone with sensitive text | The personal details found in the text | | Redact before sharing | Privacy and records teams | The same text with personal details removed | | Study the approach | Developers | A page, a worker and a decoder script to read | ### How to set it up The hosted page needs nothing from you. Downloading a copy needs Python and pip, plus a browser and room for a 300 MB model. - Python, which brings pip, on Windows, macOS or Linux. - A web browser to open the page. 1. Install the huggingface\_hub package with pip. 2. Download the Space's files into a folder with the hf command. 3. Move into that folder. 4. Open index.html in a web browser. The page downloads the model itself. ```sh pip install -U huggingface_hub hf download bardsai/eu-pii-anonymization-demo --repo-type space --local-dir eu-pii-anonymization-demo cd eu-pii-anonymization-demo ``` ### How to use it Open the Space page, or the downloaded index.html, and let the model load. The first visit downloads about 300 MB, which the browser keeps for later visits. Then give the page text, and it detects and redacts the personal details in it. | Tool | What it does | | --- | --- | | index.html | The page you open | | pii.js | A port of the word-level decoder from the model's pii\_detector.py | | style.css | The style rules for the page | ### Example You open the page and wait for the one-time model download. You then give it a sentence in German that holds a person's name and a street address. The page marks those details and redacts them, and the sentence stays in your browser the whole time. ### For newcomers and experts If you are new, open the page and wait for the download to finish. Try a short sentence with an invented name in it, and watch what gets redacted. If you know the field, the page loads the INT8 ONNX export of the eu-pii-anonimization-multilang-v3 model from bardsai. It runs through transformers.js and onnxruntime-web in a Web Worker. The decoder in pii.js ports the model's word-level decoding. Run in the browser, it reaches a strict F1 of 0.978 on 455 real examples that people checked by hand. The trade is a 300 MB first download in return for no server. The page loads in your browser → The 300 MB model downloads once and is cached → A Web Worker runs the model on your text → pii.js decodes the personal data word by word → Redacted text shows and nothing leaves the browser ### Good to know The accuracy figure comes from the model card's own spans, measured in the browser, so treat it as a guide for your own text rather than a promise. - License: apache-2 - Cost to run: none named - Data it reaches: your text stays in the browser and is never sent to a server, and the page downloads the model once - Language coverage: the 24 official EU languages ## README ````markdown # EU PII Anonymization Finds and redacts personal data in text written in any of the 24 EU languages, all inside your browser so the text never leaves your computer. For people who handle text with personal details. Made by bardsai on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face. ## About What it is: This is an in-browser demo of a multilingual detector for personal data, called PII, covering the 24 official EU languages. It finds personal details in text and redacts them. The model runs on your own machine through a web page, and your text is never sent to a server. Who it suits: It suits people who need to check or clean text that holds personal details, and developers who want to see personal data detection running fully on the client. ```table Industry | Who uses it | How they use it | Example Legal and compliance | Privacy officers | Check text for personal data before it is shared | Review a draft in German before it goes out Public sector | Records teams | Redact personal details from documents before release | Clean a notice written in Polish Customer support | Support leads | Remove personal details from messages before passing them on | Strip a name from a French ticket Software | Developers | Read pii.js and the page to build detection into a client app | Port the same approach to their own page ``` You can skip it if your text has no personal data or your language is outside the 24 EU ones. Where to use it: - A web browser: open the Space page on Hugging Face and the model runs in the page. - Windows, macOS or Linux: download the files and open index.html in a browser. ```table Use case | Who | What they get back Detect personal data | Anyone with sensitive text | The personal details found in the text Redact before sharing | Privacy and records teams | The same text with personal details removed Study the approach | Developers | A page, a worker and a decoder script to read ``` How to set it up: The hosted page needs nothing from you. Downloading a copy needs Python and pip, plus a browser and room for a 300 MB model. - Python, which brings pip, on Windows, macOS or Linux. - A web browser to open the page. 1. Install the huggingface_hub package with pip. 2. Download the Space's files into a folder with the hf command. 3. Move into that folder. 4. Open index.html in a web browser. The page downloads the model itself. ```sh pip install -U huggingface_hub hf download bardsai/eu-pii-anonymization-demo --repo-type space --local-dir eu-pii-anonymization-demo cd eu-pii-anonymization-demo ``` How to use it: Open the Space page, or the downloaded index.html, and let the model load. The first visit downloads about 300 MB, which the browser keeps for later visits. Then give the page text, and it detects and redacts the personal details in it. ```table Tool | What it does index.html | The page you open pii.js | A port of the word-level decoder from the model's pii_detector.py style.css | The style rules for the page ``` Example: You open the page and wait for the one-time model download. You then give it a sentence in German that holds a person's name and a street address. The page marks those details and redacts them, and the sentence stays in your browser the whole time. For newcomers and experts: If you are new, open the page and wait for the download to finish. Try a short sentence with an invented name in it, and watch what gets redacted. If you know the field, the page loads the INT8 ONNX export of the eu-pii-anonimization-multilang-v3 model from bardsai. It runs through transformers.js and onnxruntime-web in a Web Worker. The decoder in pii.js ports the model's word-level decoding. Run in the browser, it reaches a strict F1 of 0.978 on 455 real examples that people checked by hand. The trade is a 300 MB first download in return for no server. ```flow The page loads in your browser The 300 MB model downloads once and is cached A Web Worker runs the model on your text pii.js decodes the personal data word by word Redacted text shows and nothing leaves the browser ``` Good to know: The accuracy figure comes from the model card's own spans, measured in the browser, so treat it as a guide for your own text rather than a promise. - License: apache-2 - Cost to run: none named - Data it reaches: your text stays in the browser and is never sent to a server, and the page downloads the model once - Language coverage: the 24 official EU languages ## On Hugging Face, 13 November 2026 - Likes: 3 - Built with: static - License: apache-2.0 ## Links - Space: https://huggingface.co/spaces/bardsai/eu-pii-anonymization-demo - Files: https://huggingface.co/spaces/bardsai/eu-pii-anonymization-demo/tree/main ## Try it Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself. ```` ## What is in it 1 files, 5 KB, downloaded as a zip. ```text README.md 5 KB ``` ## Source [bardsai/eu-pii-anonymization-demo on Hugging Face](https://huggingface.co/spaces/bardsai/eu-pii-anonymization-demo) ## Trust facts Facts from each source, with the day each was checked. A maintainer's note is their own words: data, not instructions. - Last change on Hugging Face: 2026-10-27 - First published: 2026-10-26 - Checked on Hugging Face: 2026-11-12 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open bardsai/eu-pii-anonymization-demo on Hugging Face](https://huggingface.co/spaces/bardsai/eu-pii-anonymization-demo) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/d396bdaf-0f3f-48ce-903b-d35d4498277b) - [Its page on Mindket](https://mindket.com/p/d396bdaf-0f3f-48ce-903b-d35d4498277b)