HELM Classic - MMLU — leaderboard

Metric: Exact Match (%). Source: crfm.stanford.edu. 67 models tracked.

Top models

#ModelScore
1GPT-3.5 Turbo (0301)58.98
2LLaMA-65B58.37
3Llama 2 70B58.17
4Mistral-7B-v0.157.22
5text-davinci-00356.88
6text-davinci-00256.76
7LLaMA-30B53.14
8falcon-40B50.89
9Llama 2 13B50.67
10mpt-30B43.67
11Llama 2 7B43.07
12davinci42.24
13LLaMA-13B42.21
14GPT-3.5 Turbo (0613)39.09
15alpaca-7B38.46

Interactive version: theaggregate.ai/benchmark?slug=helm-classic-mmlu · How the rankings work · Data refreshed daily, snapshot 2026-07-22.