HELM Arabic - Mbzuai Human Translated Arabic Mmlu - Anatomy: leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 83.7 |
| 2 | GPT-5.4 (2026-03-05) | 81.48 |
| 3 | Claude Sonnet 4.6 | 80.74 |
| 4 | GPT-4.1 (2025-04-14) | 77.04 |
| 5 | Gemma 4 31B (IT) | 73.33 |
| 6 | Qwen 3.5 397B A17B | 73.33 |
| 7 | Llama 4 Maverick Instruct FP8 | 71.85 |
| 8 | Gemini 2.5 Flash (Thinking) | 71.85 |
| 9 | Claude Haiku 4.5 (20251001) | 68.15 |
| 10 | Mistral Large 3 | 66.67 |
| 11 | Gemini 2.5 Flash Lite (Thinking) | 65.93 |
| 12 | GPT-4.1 Mini | 64.44 |
| 13 | Qwen 3 Next 80B A3B Instruct | 62.96 |
| 14 | DeepSeek V3.1 | 62.22 |
| 15 | Llama 4 Scout Instruct | 61.48 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-mbzuai-human-translated-arabic-mmlu-anatomy · How It Works · Data refreshed daily, snapshot 2026-09-19.