MEDIC Benchmark: leaderboard
Clinical LLM benchmark leaderboard spanning closed-ended medical QA, open-ended clinical tasks, medical safety, summarization, note generation, HealthBench, EHRSQL, MedCalc, MedEC, general-domain, and DischargeMe evaluations.
Metric: MEDIC Public Table Average. Source: huggingface.co. Source: benchmarklist.com. Saturation forecast: Not forecast. 96 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash (Preview 04-17) (Thinking) | 92.36 | #237 (Gemini 2.5 Flash (Preview 04-17)) |
| 2 | GPT-4.1 | 91.71 | #248 |
| 3 | O4 Mini | 90.5 | #180 |
| 4 | Qwen 2 72B Instruct | 85.87 | #576 |
| 5 | QwQ 32B-Preview | 85.56 | #651 |
| 6 | Qwen 2.5 72B | 85.46 | #499 |
| 7 | Qwen 3 30B A3B | 84.38 | #493 |
| 8 | gemma-2-9B-it-SimPO | 81.55 | #580 |
| 9 | Qwen 2.5 7B | 81.31 | #922 |
| 10 | Falcon3-10B-Instruct | 81.03 | #1157 |
| 11 | Falcon3-7B-Instruct | 79.83 | #1305 |
| 12 | Qwen 2.5 3B | 78.11 | #1211 |
| 13 | Llama 3.1 405B Instruct | 76.81 | #457 |
| 14 | Falcon3-3B-Instruct | 75.28 | #1456 |
| 15 | Kimi K2 (Thinking) | 74.36 | #243 (Kimi K2) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medic-benchmark · How It Works · Data refreshed daily, snapshot 2026-10-20.