Open LLM Leaderboard - MMLU-Pro: leaderboard
The "MMLU-Pro" subtask of Open LLM Leaderboard. A task from the Open LLM Leaderboard, the standard open-weight evaluation suite spanning knowledge, reasoning, math and truthfulness benchmarks.
Metric: Score. Source: huggingface.co. Saturation forecast: Estimated already saturated. 4576 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | calme-3.2-instruct-78B | 73.03 | #347 |
| 2 | Qwen 2.5 72B | 59.68 | #499 |
| 3 | Rombos-LLM-V2.5-Qwen-32B | 59.16 | #367 |
| 4 | oxyge1-33B | 59.09 | #690 |
| 5 | Qwentile2.5-32B-Instruct | 58.79 | #339 |
| 6 | Qwen 2.5 32B | 58.05 | #546 |
| 7 | Qwen 2 72B | 57.31 | #521 |
| 8 | Awqward2.5-32B-Instruct | 57.23 | #371 |
| 9 | QwQ 32B-Preview | 56.78 | #651 |
| 10 | Qwen 2.5 32B Instruct | 56.67 | #498 |
| 11 | Qwen 2.5 72B Instruct | 56.26 | #454 |
| 12 | Qwen2.5-32B-Instruct-abliterated-v2 | 56.22 | #394 |
| 13 | Mistral Large 2 (Nov) Instruct (2411) | 55.62 | #465 |
| 14 | Mistral-Small-24B-Base-2501 | 54.06 | #670 |
| 15 | Qwen 2 72B Instruct | 54.03 | #576 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-llm-leaderboard-mmlu-pro · How It Works · Data refreshed daily, snapshot 2026-10-20.