HELM TORR - Qtsumm - Robustness: leaderboard
Metric: Robustness (0-100). Source: crfm.stanford.edu. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-11-20) | 79.31 |
| 2 | Claude 3.5 Sonnet (20241022) | 78.96 |
| 3 | GPT-4o Mini (2024-07-18) | 78.71 |
| 4 | Qwen 2 72B Instruct | 77.49 |
| 5 | DeepSeek V3 | 77.42 |
| 6 | Claude 3.5 Haiku (20241022) | 76.93 |
| 7 | Llama 3.1 70B Instruct | 75.39 |
| 8 | Gemini 1.5 Pro (002) | 73.47 |
| 9 | Llama 3.1 405B Instruct | 72.85 |
| 10 | Mistral 7B Instruct (v0.3) | 72.09 |
| 11 | Llama 3.1 8B Instruct | 65.76 |
| 12 | Gemini 1.5 Flash (002) | 60.32 |
Interactive version: theaggregate.ai/benchmark?slug=helm-torr-qtsumm-robustness · How It Works · Data refreshed daily, snapshot 2026-09-19.