HELM Arabic - Arabic Mmlu - Islamic Studies: leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 92.96 |
| 2 | Gemini 2.5 Flash (Thinking) | 88.73 |
| 3 | Claude Sonnet 4.6 | 88.58 |
| 4 | GPT-4.1 (2025-04-14) | 88.42 |
| 5 | GPT-5.4 (2026-03-05) | 88.42 |
| 6 | Mistral Large 3 | 88.11 |
| 7 | Gemma 4 31B (IT) | 83.72 |
| 8 | Qwen 3 Next 80B A3B Instruct | 80.59 |
| 9 | Claude Haiku 4.5 (20251001) | 80.44 |
| 10 | Llama 4 Maverick Instruct FP8 | 80.13 |
| 11 | Qwen 3.5 397B A17B | 79.81 |
| 12 | DeepSeek V3.1 | 79.19 |
| 13 | Gemini 2.5 Flash Lite (Thinking) | 76.21 |
| 14 | ALLaM-7B-Instruct-preview | 75.43 |
| 15 | AceGPT-v2-70B-Chat | 72.46 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-arabic-mmlu-islamic-studies · How It Works · Data refreshed daily, snapshot 2026-09-19.