HELM TORR - Scigen - Robustness: leaderboard
Metric: Robustness (0-100). Source: crfm.stanford.edu. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet (20241022) | 93.23 |
| 2 | GPT-4o Mini (2024-07-18) | 92.77 |
| 3 | DeepSeek V3 | 92.56 |
| 4 | Llama 3.1 8B Instruct | 92.19 |
| 5 | Claude 3.5 Haiku (20241022) | 91.81 |
| 6 | GPT-4o (2024-11-20) | 91.76 |
| 7 | Gemini 1.5 Flash (002) | 89.93 |
| 8 | Llama 3.1 70B Instruct | 88.79 |
| 9 | Qwen 2 72B Instruct | 87.71 |
| 10 | Gemini 1.5 Pro (002) | 87.3 |
| 11 | Mistral 7B Instruct (v0.3) | 86.85 |
| 12 | Llama 3.1 405B Instruct | 85.19 |
Interactive version: theaggregate.ai/benchmark?slug=helm-torr-scigen-robustness · How It Works · Data refreshed daily, snapshot 2026-09-19.