Share your thoughts, 1 month free Claude Pro on usSee more

Natural Language Understanding on AGIEval

71.6Accuracy

Llama 3 405B

Updated 8d ago

Evaluation Results

Method	Links
Llama 3 405B 2024.07		71.6
Llama 3 70B 2024.07		64.6
Mixtral 8x22B 2024.07		61.5
Qwen2.5 2025.12		48.13
Llama 3 8B 2024.07		47.8
Gemma 7B 2024.07		46
Mistral 7B 2024.07		42.7
Qwen3 2025.12		42.14
NITP 2026.05		35.11
Gamayun 2025.12		33.34
mPLUG-Owl2 2023.11		32.7
NTP 2026.05		31.77
NITP 2026.05		31.66
NTP 2026.05		30.51
Arcana 2024.10		29.3
LLaMA-2-Chat 2023.11		28.5
LLaMA-2-Chat 2024.10		28.5
Gemma3 2025.12		28.05
DeepSeek-VL 2024.03		27.8
NTP 2026.05		27.01
Llama3.2 2025.12		26.32
NITP 2026.05		26.15
WizardLM 2023.11		23.2
WizardLM 2024.10		23.2
LLaMA-2 2023.11		21.8
LLaMA-2 2024.10		21.8
Vicuna-v1.5 2023.11		21.2
Vicuna-v1.5 2024.10		21.2
DeepSeek-LLM 2024.03		19.3
DeepSeek-VL 2024.03		14