HELM MMLU - Moral Scenarios: leaderboard
The "Moral Scenarios" 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 | GPT-4 (0613) | 90.17 | #494 |
| 2 | Claude 3.5 Sonnet (20241022) | 88.83 | #291 |
| 3 | Claude 3.5 Sonnet (20240620) | 88.16 | #298 |
| 4 | Llama 3.1 405B Instruct | 87.6 | #461 |
| 5 | GPT-4o (2024-05-13) | 84.13 | #326 |
| 6 | Llama 3.2 90B Vision Instruct | 84.13 | #626 |
| 7 | Mistral Large 2 (Jul) | 83.91 | #485 |
| 8 | Llama 3.1 70B Instruct | 83.35 | #566 |
| 9 | Claude 3 Opus (20240229) | 82.57 | #413 |
| 10 | GPT-4 Turbo (Preview) | 81.56 | #399 |
| 11 | Qwen 2 72B Instruct | 81.45 | #579 |
| 12 | Gemini 2.0 Flash (Preview) | 81.45 | #366 |
| 13 | DeepSeek V3 | 80.78 | #323 |
| 14 | GPT-4 Turbo | 80.34 | #406 |
| 15 | GPT-4o (2024-08-06) | 80.22 | #335 |
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-moral-scenarios · How It Works · Data refreshed daily, snapshot 2026-10-23.