BlueBench - Knowledge — leaderboard

Metric: Score (%). Source: huggingface.co. 18 models tracked.

Top models

#ModelScore
1O171.43
2GPT-4.158.16
3Mistral Large55.1
4Llama 3.3 70B Instruct53.06
5O3 Mini50
6GPT-4o45.92
7GPT-4.1 Nano44.9
8Mistral Medium 344.9
9GPT-4.1 Mini42.86
10O4 Mini40.82
11Pixtral-12B38.78
12Llama 3.2 3B Instruct35.71
13Llama 3.2 1B Instruct19.39

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