aiXamine - Model & Data Privacy: leaderboard
Metric: Service score (0-100; normalized pass rate). Source: arxiv.org. 42 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 91.1 |
| 2 | Llama 3.2 3B | 88.1 |
| 3 | Llama 3.2 1B | 86.2 |
| 4 | DeepSeek R1 0528 | 84.7 |
| 5 | Phi-4 | 84.2 |
| 6 | Claude 3.5 Haiku | 80.8 |
| 7 | Claude Haiku 4.5 | 78.7 |
| 8 | Llama 3.1 8B | 78.6 |
| 9 | GPT-OSS-120B | 76.9 |
| 10 | Llama 4 Scout | 74.5 |
| 11 | DeepSeek R1 Distill Qwen 14B | 74.4 |
| 12 | Llama 3.3 70B | 74.4 |
| 13 | DeepSeek R1 Distill Llama 8B | 73.5 |
| 14 | DeepSeek R1 0528 Qwen3 8B | 73.4 |
| 15 | Gemma 2 27B | 72.2 |
Interactive version: theaggregate.ai/benchmark?slug=aixamine-model-data-privacy · How It Works · Data refreshed daily, snapshot 2026-09-19.