Text-only Medical Question Answering on EHRNoteQA
95.15AccuracyGPT-4
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4Model Type=External SOTA2026.04 | 95.15 | |
| Gemini 3.0 ProModel Type=Large Models2026.04 | 95 | |
| Gemini 3.0 FlashModel Type=Large Models2026.04 | 93.9 | |
| MedGemma 1 27BModel Type=Small Models2026.04 | 90.7 | |
| Qwen3 VL 4BModel Type=Small Models2026.04 | 90.6 | |
| Gemma 3 27BModel Type=Small Models2026.04 | 90.3 | |
| MedGemma 1.5 4BModel Type=Small Models2026.04 | 80.4 | |
| MedGemma 1 4BModel Type=Small Models2026.04 | 79.4 | |
| Gemma 3 4BModel Type=Small Models2026.04 | 78 |