gpt-oss-safeguard-20b
Availablegpt-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
- 131.1K tokens
- Max output
- 65.5K tokens
- Input / 1M tokens
- $0per 1M tokens
- Output / 1M tokens
- $0per 1M tokens
- Listed since
- Oct 29, 2025
Prices and model data from OpenRouter
Where it's availableProviders · 2
| Provider | Input / 1M tokens | Output / 1M tokens | Status |
|---|---|---|---|
| Groq | Free | Free | Available |
| OpenRouter | $0.075 | $0.30 | Available |
Published benchmark results
From independent leaderboards, each credited to its publisher.