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

Multitask Language Understanding on MMLU Medical Genetics FR (test)

86Accuracy

MedGemma-27B-it

Updated 9d ago

Evaluation Results

Method	Links
MedGemma-27B-it 2026.04		86
Qwen3-32B 2026.04		85
Qwen3-8B-Base 2026.04		80
Qwen3-8B 2026.04		80
Qwen3-14B 2026.04		80
Qwen3-4B-Base 2026.04		74
Qwen3-4B 2026.04		69
Llama3.1-8B-Instruct 2026.04		68
Llama3.2-1B 2026.04		67
Llama3.1-8B 2026.04		67
Apollo-7B 2026.04		67
Qwen3-1.7B-Base 2026.04		66
Gaperon-8B 2026.04		65
MedGemma-4B-it 2026.04		65
Apertus-8B 2026.04		64
EuroLLM-9B-Instruct 2026.04		64
GPT-OSS-20B 2026.04		64
Olmo3-7B 2026.04		63
EuroLLM-9B 2026.04		62
Qwen3-1.7B 2026.04		62
Apertus-8B-Instruct 2026.04		62
Mistral-7B-Instruct-v0.3 2026.04		60
Mistral-7B-v0.3 2026.04		59
SmolLM-3B 2026.04		58
MedGemma-4B-pt 2026.04		48
BioMistral-7B 2026.04		48
Qwen3-0.6B-Base 2026.04		47
Llama3.2-1B-Instruct 2026.04		46
Gemma3-1B-it 2026.04		44
Qwen3-0.6B 2026.04		38
Gemma3-270m-it 2026.04		28
Gaperon-1B 2026.04		27
Gemma3-1B-pt 2026.04		26
Baguettotron 2026.04		24
Gemma3-270m-pt 2026.04		23