MMTU - KB mapping — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 26 models tracked.

Top models

#ModelScore
1O361.44
2GPT-561.18
3Gemini 2.5 Pro60.84
4Grok 3 Mini54.8
5GPT-5 Chat54.27
6GPT-5 Mini52.77
7GPT-4o (2024-11-20)49.84
8DeepSeek R1 052849
9Qwen 3 235B A22B (Thinking)47.63
10Gemini 2.5 Flash47.16
11Mistral Large 2 (Nov) Instruct (2411)45.67
12Llama 3.3 70B43.55
13GPT-OSS-120B42.32
14Llama 4 Maverick Instruct FP842.02
15Qwen 3 235B A22B 2507 Instruct41.29

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