OpenAI launches open-source security inference model gpt-oss-safeguard, supporting policy-driven classification.

This model, fine-tuned based on open-weight gpt-oss and licensed under the Apache 2.0 license, is available for download from Hugging Face. Internal evaluations show that it outperforms gpt-5-thinking and gpt-oss in multi-policy accuracy, and its performance on external datasets is close to that of Safety Reasoner.