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

Multitask Language Understanding on MMLU Professional Medicine FR (test)

84.19Accuracy

Qwen3-32B

Updated 9d ago

Evaluation Results

Method	Links
Qwen3-32B 2026.04		84.19
Qwen3-14B 2026.04		78.68
MedGemma-27B-it 2026.04		78.68
Qwen3-8B-Base 2026.04		76.47
Qwen3-8B 2026.04		72.79
Qwen3-4B-Base 2026.04		69.49
Qwen3-4B 2026.04		68.38
Llama3.1-8B-Instruct 2026.04		61.4
Apertus-8B-Instruct 2026.04		60.66
Apollo-7B 2026.04		59.93
Apertus-8B 2026.04		59.56
Qwen3-1.7B-Base 2026.04		56.25
EuroLLM-9B-Instruct 2026.04		56.25
EuroLLM-9B 2026.04		55.88
Llama3.2-1B 2026.04		55.51
Llama3.1-8B 2026.04		55.51
Mistral-7B-v0.3 2026.04		54.41
Mistral-7B-Instruct-v0.3 2026.04		52.94
GPT-OSS-20B 2026.04		52.94
MedGemma-4B-it 2026.04		51.47
SmolLM-3B 2026.04		49.26
Olmo3-7B 2026.04		48.9
Qwen3-1.7B 2026.04		48.16
BioMistral-7B 2026.04		46.32
Gaperon-8B 2026.04		42.28
Qwen3-0.6B-Base 2026.04		37.13
MedGemma-4B-pt 2026.04		35.29
Qwen3-0.6B 2026.04		33.46
Gaperon-1B 2026.04		30.88
Llama3.2-1B-Instruct 2026.04		26.47
Gemma3-1B-pt 2026.04		24.63
Gemma3-270m-pt 2026.04		23.16
Baguettotron 2026.04		20.96
Gemma3-1B-it 2026.04		20.96
Gemma3-270m-it 2026.04		18.75