OpenAI: gpt-oss-safeguard-20b

gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust...

• Context window: 131,072 tokens
• Input: text
• Output: text
• Pricing: $0.075/M input tokens, $0.3/M output tokens

View on OpenRouter. Model data sourced from OpenRouter.