HELM TORR - Tab Fact: leaderboard
Metric: Accuracy. Source: crfm.stanford.edu. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet (20241022) | 85.54 |
| 2 | GPT-4o (2024-11-20) | 83.09 |
| 3 | Llama 3.1 405B Instruct | 82.31 |
| 4 | DeepSeek V3 | 81 |
| 5 | Gemini 1.5 Pro (002) | 79.57 |
| 6 | Claude 3.5 Haiku (20241022) | 78.91 |
| 7 | Qwen 2 72B Instruct | 78.37 |
| 8 | Gemini 1.5 Flash (002) | 76.83 |
| 9 | Llama 3.1 70B Instruct | 72.91 |
| 10 | GPT-4o Mini (2024-07-18) | 65.09 |
| 11 | Mistral 7B Instruct (v0.3) | 55.46 |
| 12 | Llama 3.1 8B Instruct | 55.03 |
Interactive version: theaggregate.ai/benchmark?slug=helm-torr-tab-fact · How It Works · Data refreshed daily, snapshot 2026-09-08.