HELM Arabic - Arabic Mmlu - Computer Science (University): leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 (2025-04-14) | 85.94 |
| 2 | Claude Opus 4.7 | 85.94 |
| 3 | Claude Sonnet 4.6 | 84.38 |
| 4 | Gemma 4 31B (IT) | 84.38 |
| 5 | GPT-5.4 (2026-03-05) | 84.38 |
| 6 | Llama 4 Scout Instruct | 82.81 |
| 7 | Qwen 3.5 397B A17B | 82.81 |
| 8 | Gemini 2.5 Flash (Thinking) | 82.81 |
| 9 | Llama 3.3 70B Instruct | 81.25 |
| 10 | Qwen 3 Next 80B A3B Instruct | 81.25 |
| 11 | Mistral Large 3 | 81.25 |
| 12 | GPT-4.1 Mini | 79.69 |
| 13 | AceGPT-v2-70B-Chat | 79.69 |
| 14 | Qwen 2.5 72B Instruct | 78.12 |
| 15 | DeepSeek V3.1 | 78.12 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-arabic-mmlu-computer-science-university · How It Works · Data refreshed daily, snapshot 2026-09-19.