HELM MMLU - Prehistory: leaderboard
The "Prehistory" 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) | 95.06 | #298 |
| 2 | Claude 3.5 Sonnet (20241022) | 94.14 | #291 |
| 3 | Claude 3 Opus (20240229) | 94.14 | #413 |
| 4 | GPT-4o (2024-05-13) | 93.83 | #326 |
| 5 | GPT-4o (2024-08-06) | 93.52 | #335 |
| 6 | Llama 3.1 405B Instruct | 92.9 | #461 |
| 7 | Gemini 1.5 Pro (002) | 92.59 | #350 |
| 8 | GPT-4 (0613) | 92.59 | #494 |
| 9 | Nova Pro | 92.59 | #573 |
| 10 | DeepSeek V3 | 92.28 | #323 |
| 11 | GPT-4 Turbo | 91.98 | #406 |
| 12 | Mistral Large 2 (Jul) | 91.98 | #485 |
| 13 | GPT-4 Turbo (Preview) | 91.67 | #399 |
| 14 | Qwen 2 72B Instruct | 91.36 | #579 |
| 15 | Qwen 2.5 72B Instruct | 91.05 | #458 |
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-prehistory · How It Works · Data refreshed daily, snapshot 2026-10-23.