HELM Arabic - Arabic Mmlu - General Knowledge (Middle School): leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 31B (IT) | 86.63 |
| 2 | Claude Opus 4.7 | 86.63 |
| 3 | Claude Sonnet 4.6 | 86.05 |
| 4 | Qwen 3.5 397B A17B | 84.88 |
| 5 | Mistral Large 3 | 84.88 |
| 6 | GPT-5.4 (2026-03-05) | 84.88 |
| 7 | Qwen 2.5 72B Instruct | 83.72 |
| 8 | Gemini 2.5 Flash (Thinking) | 83.72 |
| 9 | GPT-4.1 (2025-04-14) | 83.14 |
| 10 | Qwen 3 Next 80B A3B Instruct | 83.14 |
| 11 | Claude Haiku 4.5 (20251001) | 82.56 |
| 12 | Gemini 2.5 Flash Lite (Thinking) | 82.56 |
| 13 | Llama 4 Maverick Instruct FP8 | 81.98 |
| 14 | Llama 4 Scout Instruct | 79.65 |
| 15 | Qwen 3.5 9B | 79.65 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-arabic-mmlu-general-knowledge-middle-school · How It Works · Data refreshed daily, snapshot 2026-09-19.