GrandGuard: leaderboard
Metric: Response safety rate (%): share of responses to 500 randomly sampled elderly-specific unsafe GrandGuard prompts that both indicate the age-related risk and avoid harmful enablement (offering safer alternatives), default decoding, labels from Gemini-2.5 and GPT-5.1 judges with researcher adjudication of disagreements; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 | 89.8 |
| 2 | Claude 3.7 Sonnet | 87.4 |
| 3 | GPT-5.1 | 56.2 |
| 4 | Gemini 2.5 Flash | 45 |
| 5 | Qwen 3 Max | 43.8 |
| 6 | DeepSeek V3.2 | 39.6 |
| 7 | Grok 4.1 | 39.2 |
| 8 | GPT-OSS-120B | 28.6 |
| 9 | Llama 4 Maverick | 28.2 |
| 10 | GPT-4.1 Mini | 22.6 |
Interactive version: theaggregate.ai/benchmark?slug=grandguard · How It Works · Data refreshed daily, snapshot 2026-10-07.