---
title: privacy-filter
canonical_url: "https://www.modelscope.cn/studios/AI-ModelScope/privacy-filter"
md_url: "https://www.modelscope.cn/studios/AI-ModelScope/privacy-filter.md"
repository: AI-ModelScope/privacy-filter
chinese_name: "OpenAI隐私过滤模型"
last_updated: 2026-06-16
sdk_type: gradio
sdk_version: 6.2.0
downloads: 0
stars: 21
---

# privacy-filter

> privacy-filter - AI-ModelScope 在 ModelScope 创建的在线 Demo。OpenAI Privacy Filter

AI-ModelScope/privacy-filter 是 ModelScope 魔搭社区上的在线可交互 Demo（创空间），基于 gradio 6.2.0 构建，中文名为「OpenAI隐私过滤模型」。

- **Repository**: AI-ModelScope/privacy-filter
- **SDK**: gradio
- **SDK version**: 6.2.0
- **Downloads**: 0
- **Stars**: 21
- **Last updated**: 2026-06-16

Source: https://www.modelscope.cn/studios/AI-ModelScope/privacy-filter

---

# OpenAI Privacy Filter

OpenAI Privacy Filter is a bidirectional token-classification model for personally identifiable information (PII) detection and masking in text. It is intended for high-throughput data sanitization workflows where teams need a model that they can run on-premises that is fast, context-aware, and tunable.

OpenAI Privacy Filter is pretrained autoregressively to arrive at a checkpoint with similar architecture to gpt-oss, albeit of a smaller size.  We  then converted that checkpoint into a bidirectional token classifier over a privacy label taxonomy, and post-trained with a supervised classification loss. (For architecture details about gpt-oss, please see the gpt-oss model card.) Instead of generating text token-by-token, this model labels an input sequence in a single forward pass, then decodes coherent spans with a constrained Viterbi procedure. For each input token, the model predicts a probability distribution over the label taxonomy which consists of 8 output categories described below.

Highlights:

- Permissive Apache 2.0 license: ideal for experimentation, customization, and commercial deployment.
- Small size: Runs in a web browser or on a laptop – 1.5B parameters total and 50M active parameters.
- Fine-tunable: Adapt the model to specific data distributions through easy and data efficient finetuning.
- Long-context: 128,000-token context window enables processing long text with high throughput and no chunking.
- Runtime control: configure precision/recall tradeoffs and detected span lengths through preset operating points.

## Metadata

- Developed by: OpenAI
- Funded by: OpenAI
- Shared by: OpenAI
- Model type: Bidirectional token classification model for privacy span detection
- Language(s): Primarily English; selected multilingual robustness evaluation reported
- License: [Apache 2.0](LICENSE)

- Source repository: https://github.com/openai/privacy-filter
- Model weights: https://huggingface.co/openai/privacy-filter
- Model card: [OpenAI Privacy Filter Model Card](https://cdn.openai.com/pdf/c66281ed-b638-456a-8ce1-97e9f5264a90/OpenAI-Privacy-Filter-Model-Card.pdf)