HELM Arabic - Mbzuai Human Translated Arabic Mmlu - Conceptual Physics: leaderboard
HELM Arabic (release v2.2.0) MBZUAI human-translated Arabic MMLU, Conceptual Physics: the MMLU four-option questions for this subject translated into Arabic by human translators; 235 questions, exact match. Percent correct, 0-100, higher is better; guessing scores 25.
Metric: Exact Match (%). Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 41 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 93.19 | #51 |
| 2 | Claude Sonnet 4.6 | 92.77 | #94 |
| 3 | Qwen 3.5 397B A17B | 88.09 | #154 |
| 4 | Gemini 2.5 Flash (Non-reasoning) | 86.81 | #255 (Gemini 2.5 Flash) |
| 5 | Gemma 4 31B (IT) | 85.53 | #178 |
| 6 | Claude Haiku 4.5 (20251001) | 84.26 | #264 |
| 7 | GPT-5.4 (2026-03-05) | 83.83 | #63 |
| 8 | Llama 4 Maverick Instruct FP8 | 82.13 | #424 |
| 9 | GPT-5.1 (2025-11-13) | 82.13 | #107 |
| 10 | GPT-4.1 (2025-04-14) | 81.28 | #257 |
| 11 | Qwen 3 Next 80B A3B Instruct | 81.28 | #381 |
| 12 | DeepSeek V3.1 | 81.28 | #275 |
| 13 | Qwen 3.5 9B | 80.43 | #391 |
| 14 | Mistral Large 3 | 80.43 | #410 |
| 15 | Gemini 2.5 Flash Lite (Non-reasoning) | 78.72 | #432 (Gemini 2.5 Flash Lite) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-mbzuai-human-translated-arabic-mmlu-conceptual-physics · How It Works · Data refreshed daily, snapshot 2026-11-05.