HELM Arabic - Arabic Mmlu - Social Science (Middle School): leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 64.32 |
| 2 | GPT-5.4 (2026-03-05) | 62.66 |
| 3 | Llama 3.3 70B Instruct | 62.24 |
| 4 | Mistral Large 3 | 62.24 |
| 5 | GPT-4.1 (2025-04-14) | 61.83 |
| 6 | Gemma 4 31B (IT) | 61.41 |
| 7 | Llama 4 Maverick Instruct FP8 | 61.41 |
| 8 | Gemini 2.5 Flash (Thinking) | 61.41 |
| 9 | Llama 4 Scout Instruct | 61 |
| 10 | Qwen 3 Next 80B A3B Instruct | 60.58 |
| 11 | Qwen 3.5 397B A17B | 60.58 |
| 12 | Claude Sonnet 4.6 | 60.17 |
| 13 | Qwen 2.5 72B Instruct | 59.34 |
| 14 | DeepSeek V3.1 | 59.34 |
| 15 | Claude Haiku 4.5 (20251001) | 58.92 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-arabic-mmlu-social-science-middle-school · How It Works · Data refreshed daily, snapshot 2026-09-19.