BlueBench - Legal — leaderboard

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

Top models

#ModelScore
1Mistral Medium 367.17
2GPT-4o60.87
3GPT-4.160.47
4Llama 3.3 70B Instruct59.54
5GPT-4.1 Mini59.32
6GPT-4.1 Nano56.91
7O4 Mini56.4
8O151.22
9Llama 3.2 3B Instruct50.56
10O3 Mini37.87
11Llama 3.2 1B Instruct37.26
12Mistral Large34.71
13Pixtral-12B32.9

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