HELM Arabic - Arabic Mmlu - Computer Science (Middle School): leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 4 Scout Instruct | 100 |
| 2 | GPT-4.1 (2025-04-14) | 100 |
| 3 | Gemma 4 31B (IT) | 100 |
| 4 | Mistral Large 2 (Nov) Instruct (2411) | 100 |
| 5 | GPT-4.1 Mini | 96.3 |
| 6 | Llama 3.3 70B Instruct | 96.3 |
| 7 | Qwen 3.5 397B A17B | 96.3 |
| 8 | Mistral Large 3 | 96.3 |
| 9 | Gemini 2.5 Flash (Thinking) | 96.3 |
| 10 | Qwen 2.5 72B Instruct | 92.59 |
| 11 | Claude Opus 4.7 | 92.59 |
| 12 | Claude Haiku 4.5 (20251001) | 92.59 |
| 13 | Qwen 3 Next 80B A3B Instruct | 92.59 |
| 14 | DeepSeek V3.1 | 92.59 |
| 15 | Gemini 2.5 Flash Lite (Thinking) | 92.59 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-arabic-mmlu-computer-science-middle-school · How It Works · Data refreshed daily, snapshot 2026-09-19.