Accuracy on MMLU-Pro Health (EN) X (test)
68.85Accuracy (%)Qwen3-32B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-32BType=Instruct2026.04 | 68.85 | |
| MedGemma-27B-itType=BioMed2026.04 | 65.36 | |
| Qwen3-14BType=Instruct2026.04 | 63.03 | |
| Qwen3-8BType=Instruct2026.04 | 57.5 | |
| Qwen3-8B-BaseType=Base2026.04 | 55.75 | |
| Qwen3-4BType=Instruct2026.04 | 52.11 | |
| Qwen3-4B-BaseType=Base2026.04 | 49.93 | |
| Llama3.1-8B-InstructType=Instruct2026.04 | 49.64 | |
| Apollo-7BType=BioMed2026.04 | 43.67 | |
| Llama3.1-8BType=Base2026.04 | 43.38 | |
| Apertus-8B-InstructType=Instruct2026.04 | 41.48 | |
| MedGemma-4B-itType=BioMed2026.04 | 41.48 | |
| Mistral-7B-Instruct-v0.3Type=Instruct2026.04 | 40.9 | |
| GPT-OSS-20BType=Instruct2026.04 | 37.26 | |
| SmolLM-3BType=Instruct2026.04 | 36.97 | |
| EuroLLM-9B-InstructType=Instruct2026.04 | 36.97 | |
| Mistral-7B-v0.3Type=Base2026.04 | 35.52 | |
| Qwen3-1.7BType=Instruct2026.04 | 35.52 | |
| Olmo3-7BType=Base2026.04 | 34.93 | |
| Qwen3-1.7B-BaseType=Base2026.04 | 34.64 | |
| Apertus-8BType=Base2026.04 | 34.64 | |
| EuroLLM-9BType=Base2026.04 | 33.92 | |
| BioMistral-7BType=BioMed2026.04 | 32.17 | |
| Gaperon-8BType=Base2026.04 | 26.64 | |
| Qwen3-0.6B-BaseType=Base2026.04 | 22.42 | |
| Qwen3-0.6BType=Instruct2026.04 | 22.27 | |
| Llama3.2-1B-InstructType=Instruct2026.04 | 22.13 | |
| MedGemma-4B-ptType=BioMed2026.04 | 20.67 | |
| Gemma3-1B-itType=Instruct2026.04 | 16.16 | |
| Llama3.2-1BType=Base2026.04 | 14.85 | |
| Gaperon-1BType=Base2026.04 | 12.81 | |
| BaguettotronType=Base2026.04 | 10.92 | |
| Gemma3-1B-ptType=Base2026.04 | 10.92 | |
| Gemma3-270m-itType=Instruct2026.04 | 9.9 | |
| Gemma3-270m-ptType=Base2026.04 | 9.02 |