HELM TORR - Tablebench Fact Checking - Robustness: leaderboard
Metric: Robustness (0-100). Source: crfm.stanford.edu. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet (20241022) | 60.22 |
| 2 | DeepSeek V3 | 59.75 |
| 3 | GPT-4o (2024-11-20) | 59.08 |
| 4 | Gemini 1.5 Pro (002) | 49.78 |
| 5 | Gemini 1.5 Flash (002) | 43.48 |
| 6 | Qwen 2 72B Instruct | 39.07 |
| 7 | Claude 3.5 Haiku (20241022) | 32.88 |
| 8 | GPT-4o Mini (2024-07-18) | 32 |
| 9 | Llama 3.1 405B Instruct | 21.84 |
| 10 | Llama 3.1 70B Instruct | 19.54 |
| 11 | Mistral 7B Instruct (v0.3) | 18.92 |
| 12 | Llama 3.1 8B Instruct | 9.85 |
Interactive version: theaggregate.ai/benchmark?slug=helm-torr-tablebench-fact-checking-robustness · How It Works · Data refreshed daily, snapshot 2026-09-19.