Medical Question Answering on 141-question pharmacy dataset (test)
0.92AccuracyGPT-5
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5Parameter Size=Not disclosed2025.12 | 0.92 | |
| o3Parameter Size=Not disclosed2025.12 | 0.89 | |
| Claude Opus 4.5Parameter Size=Not disclosed2025.12 | 0.87 | |
| Medical ChatParameter Size=Not disclosed2025.12 | 0.85 | |
| GPT-4oParameter Size=Not disclosed2025.12 | 0.81 | |
| o4 MiniParameter Size=Not disclosed2025.12 | 0.76 | |
| Gemini 3 (Pro)Parameter Size=Not disclosed2025.12 | 0.75 | |
| Gemini 2.0 (Flash)Parameter Size=Not disclosed2025.12 | 0.72 | |
| Gemma 3Parameter Size=27B2025.12 | 0.61 | |
| Bio-Medical Llama 3Parameter Size=8B2025.12 | 0.46 | |
| Llama 3.1Parameter Size=8B2025.12 | 0.46 |