Instructions to use openai/privacy-filter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use openai/privacy-filter with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="openai/privacy-filter")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("openai/privacy-filter", dtype="auto") - Transformers.js
How to use openai/privacy-filter with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('token-classification', 'openai/privacy-filter'); - Inference
- Notebooks
- Google Colab
- Kaggle
Running openai/privacy-filter in ChatWall β local-first PII masking for AI chats
#26
by phcollignon - opened
Hey π Just wanted to share that we ship openai/privacy-filter in ChatWall β a browser extension that
masks PII (names, emails, IBANs, keysβ¦) before your prompt reaches ChatGPT/Claude/Gemini, then restores it in
the reply.
We run it via transformers.js, 100% on-device (WebGPU + WASM fallback), only the weights come from the HF CDN, no user text ever leaves the browser. It catches the fuzzy context-dependent entities our regex can't.
Perfect fit for a privacy tool since it runs fully client-side. Thanks for releasing it π
π https://chatwall.io