gpt-oss-safeguard-20b
OpenAI · Released Oct 2025
A 20-billion-parameter open-source model from OpenAI focused on safety and content filtering with a 131k token context window.
- Strengths
- Designed to identify and mitigate unsafe content, jailbreak attempts, and policy violations with a specialized safety-oriented architecture.
- Best for
- Safeguarding applications, content moderation systems, and safety-layer implementations where detecting harmful inputs is the primary requirement.
- Limitations
- Smaller parameter count and shorter context window than OpenAI's current-generation reasoning models like GPT-5.1, making it unsuitable for complex reasoning or long-form content generation tasks.
Input / 1M
$0.075
Output / 1M
$0.3
Cached input / 1M
$0.0375
Context window
131K
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in — out —
- 90d
- in decreased 0.0% out decreased 0.0%
- 1y
- in decreased 0.0% out decreased 0.0%
- Since launch
- in decreased 0.0% out decreased 0.0%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 19 Jun 2026 | $0.075 | $0.3 | $0.0375 | Imported from OpenRouter | openrouter.ai |
| 11 Jun 2026 | $0.075 | $0.3 | $0.037 | Imported from OpenRouter | openrouter.ai |