HELM Arabic - Mbzuai Human Translated Arabic Mmlu - Virology: leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 55.42 |
| 2 | Claude Haiku 4.5 (20251001) | 53.61 |
| 3 | Qwen 3.5 397B A17B | 53.61 |
| 4 | Claude Opus 4.7 | 52.41 |
| 5 | Qwen 3 Next 80B A3B Instruct | 52.41 |
| 6 | GPT-5.4 (2026-03-05) | 52.41 |
| 7 | GPT-4.1 (2025-04-14) | 51.81 |
| 8 | AceGPT-v2-70B-Chat | 51.2 |
| 9 | Qwen 2.5 72B Instruct | 50.6 |
| 10 | Gemma 4 31B (IT) | 50 |
| 11 | Llama 4 Maverick Instruct FP8 | 50 |
| 12 | DeepSeek V3.1 | 50 |
| 13 | Gemini 2.5 Flash Lite (Thinking) | 50 |
| 14 | jais-adapted-70B-chat | 50 |
| 15 | Mistral Large 3 | 49.4 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-mbzuai-human-translated-arabic-mmlu-virology · How It Works · Data refreshed daily, snapshot 2026-09-19.