Open Arabic LLM - Arabic MMLU HT College Medicine: leaderboard
The "Arabic MMLU HT College Medicine" subtask of Open Arabic LLM. A task from the Open Arabic LLM Leaderboard, evaluating models on Arabic-language knowledge, reasoning and understanding across native and translated benchmarks.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around January 2027. 163 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 72.25 | #454 |
| 2 | AceGPT-v2-70B-Chat | 69.94 | #407 |
| 3 | Gemma 3 27B (IT) | 67.05 | #523 |
| 4 | Qwen 2.5 32B Instruct | 66.47 | #498 |
| 5 | Qwentile2.5-32B-Instruct | 66.47 | #339 |
| 6 | gemma-3-27B-pt | 65.9 | #502 |
| 7 | EZO-Qwen2.5-32B-Instruct | 64.16 | #471 |
| 8 | Awqward2.5-32B-Instruct | 64.16 | #371 |
| 9 | Qwen2.5-32B-Instruct-CFT | 64.16 | #375 |
| 10 | AIC-1 | 64.16 | #231 |
| 11 | Llama 3.3 70B Instruct | 63.58 | #534 |
| 12 | Qwen 2.5 72B | 63.01 | #499 |
| 13 | free-evo-qwen72B-v0.8-re | 62.43 | #435 |
| 14 | Gemma 2 27B (IT) | 61.85 | #645 |
| 15 | Qwen3-8B-Base | 61.85 | #626 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-arabic-mmlu-ht-college-medicine · How It Works · Data refreshed daily, snapshot 2026-10-20.