OpenAI today released the open-source model OpenAI Privacy Filter, designed to detect and redline personally identifiable information (PII) in text. The model boasts 1.5 billion total parameters and 50 million active parameters, supporting context windows of up to 128,000 tokens. Employing a bidirectional token classification model architecture, OpenAI Privacy Filter can identify eight categories of information, including private names, addresses, emails, phone numbers, URLs, dates, account names, and keys, achieving a 96% F1 score in the PII-Masking-300k benchmark. The model is currently available under the Apache 2.0 license on Hugging Face and GitHub, allowing developers to deploy and fine-tune it locally.