Medical Visual Question Answering on Harvard-FairVLMed
88.43AccuracyMedQwen
Evaluation results
| Method | Links | |
|---|---|---|
| MedQwen2026.04 | 88.43 | |
| miniGPT-Med2026.04 | 66.73 | |
| LLaVA-Med-1.52026.04 | 63.03 | |
| GPT-4o2026.04 | 61.5 | |
| Gemini-1.52026.04 | 58.53 | |
| RadFM2026.04 | 52.47 | |
| LLaVA-v1.62026.04 | 48.52 | |
| InternVL-22026.04 | 44.38 | |
| Med-Flamingo2026.04 | 42.06 | |
| Qwen-2.5-VL2026.04 | 38.06 | |
| MedVInT2026.04 | 35.92 |