HELM Arabic - Arabic Mmlu - Philosophy (High School): leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | ALLaM-7B-Instruct-preview | 82.05 |
| 2 | Claude Sonnet 4.6 | 76.92 |
| 3 | AceGPT-v2-70B-Chat | 76.92 |
| 4 | GPT-4.1 Mini | 74.36 |
| 5 | GPT-4.1 (2025-04-14) | 74.36 |
| 6 | Gemma 4 31B (IT) | 74.36 |
| 7 | Gemini 2.5 Flash Lite (Thinking) | 74.36 |
| 8 | Claude Opus 4.7 | 71.79 |
| 9 | Claude Haiku 4.5 (20251001) | 71.79 |
| 10 | DeepSeek V3.1 | 71.79 |
| 11 | Gemini 2.5 Flash (Thinking) | 71.79 |
| 12 | Qwen 2.5 72B Instruct | 69.23 |
| 13 | Qwen 3.5 397B A17B | 69.23 |
| 14 | Mistral Large 3 | 69.23 |
| 15 | jais-adapted-70B-chat | 69.23 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-arabic-mmlu-philosophy-high-school · How It Works · Data refreshed daily, snapshot 2026-09-19.