SSyncDev

MMLU-Pro

Massive Multitask Language Understanding (Pro)

Model ranking

#ModelScore (%)
1Claude 3.5 Sonnet
Anthropic
78.0%
2Gemini 1.5 Pro
Google DeepMind
75.8%
3GPT-4o
OpenAI
74.7%
4Llama 3.1 405B
Meta AI
73.3%