SLM Trustworthiness - Robustness: leaderboard
Metric: TrustLLM robustness accuracy (%; sub-task mean). Source: arxiv.org. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 8B Instruct | 71.62 |
| 2 | Qwen 2.5 1.5B Instruct | 69.38 |
| 3 | Gemma 1.1 7B (IT) | 68.9 |
| 4 | Qwen 2.5 3B Instruct | 68.79 |
| 5 | Qwen 2.5 7B Instruct | 67.05 |
| 6 | Llama 3.2 3B Instruct | 66.42 |
| 7 | Qwen 2.5 0.5B Instruct | 62.42 |
| 8 | SmolLM2-360M-Instruct | 60.16 |
| 9 | Llama 3.2 1B Instruct | 59.43 |
| 10 | h2o-danube3-500m-chat | 56.85 |
Interactive version: theaggregate.ai/benchmark?slug=slm-trustworthiness-robustness · How It Works · Data refreshed daily, snapshot 2026-09-19.