HELM MMLU - Logical Fallacies: leaderboard
The "Logical Fallacies" subtask of HELM MMLU. HELM Classic MMLU group leaderboard from the public HELM release artifacts.
Metric: EM. Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 79 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3.5 Sonnet (20240620) | 92.64 | #298 |
| 2 | Llama 3.1 405B Instruct | 92.02 | #461 |
| 3 | Claude 3.5 Sonnet (20241022) | 91.41 | #291 |
| 4 | DeepSeek V3 | 91.41 | #323 |
| 5 | Qwen 2 72B Instruct | 91.41 | #579 |
| 6 | Gemini 2.0 Flash (Preview) | 91.41 | #366 |
| 7 | Gemini 1.5 Pro (002) | 90.18 | #350 |
| 8 | GPT-4o (2024-08-06) | 90.18 | #335 |
| 9 | Claude 3 Opus (20240229) | 89.57 | #413 |
| 10 | Gemini 1.5 Pro (001) | 89.57 | #446 |
| 11 | Qwen 2.5 72B Instruct | 88.96 | #458 |
| 12 | GPT-4o (2024-05-13) | 88.34 | #326 |
| 13 | Yi 34B (Base) | 88.34 | #975 |
| 14 | Mixtral 8x22B | 87.73 | #770 |
| 15 | GPT-4o Mini (2024-07-18) | 87.12 | #561 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-mmlu-logical-fallacies · How It Works · Data refreshed daily, snapshot 2026-10-23.