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

#ModelScoreOverall rank
1calme-3.2-instruct-78B73.03#354
2Qwen 2.5 72B59.68#514
3Rombos-LLM-V2.5-Qwen-32B59.16#371
4oxyge1-33B59.09#697
5Qwentile2.5-32B-Instruct58.79#345
6Qwen 2.5 32B58.05#545
7Qwen 2 72B57.31#516
8Awqward2.5-32B-Instruct57.23#370
9Qwen 2 VL 72B Instruct57.17#551
10QwQ 32B-Preview56.78#670
11Qwen 2.5 32B Instruct56.67#511
12Qwen 2.5 72B Instruct56.26#470
13Qwen2.5-32B-Instruct-abliterated-v256.22#400
14Mistral Large 2 (Nov) Instruct (2411)55.62#481
15Mistral-Small-24B-Base-250154.06#683

No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #6 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=open-llm-leaderboard-mmlu-pro · How It Works · Data refreshed daily, snapshot 2026-11-06.