SurgiQ: leaderboard
SurgiQ (2026) is a text-only benchmark of four-option surgical multiple-choice questions across six surgical specialties, generated from textbooks, open-access papers and exam material and filtered by verification and expert audit. The score is micro accuracy in percent over all questions, reading the option label with the highest next-token probability, zero-shot; higher is better.
Metric: Micro accuracy over all 13,055 questions: accuracy (%) on SurgiQ's four-option surgical multiple-choice questions, zero-shot; the answer is the option label with the highest next-token log-probability (plain and space-prefixed labels), options shuffled once. Source: arxiv.org. Saturation forecast: Estimated already saturated. 35 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 72B | 68.08 | #1190 |
| 2 | Qwen 2.5 72B Instruct | 68 | #1092 |
| 3 | Qwen 2.5 32B | 65.43 | #1332 |
| 4 | Qwen 2.5 32B Instruct | 64.92 | #1215 |
| 5 | GPT-OSS-120B | 64.63 | #700 |
| 6 | Ministral-3-14B-Reasoning-2512 | 63.48 | #1396 |
| 7 | DeepSeek R1 Distill Qwen 32B | 60.37 | #1788 |
| 8 | Ministral-3-8B-Reasoning-2512 | 58.45 | #1983 |
| 9 | GPT-OSS-20B | 57.56 | #1147 |
| 10 | DeepSeek R1 Distill Qwen 14B | 55.92 | #2582 |
| 11 | Ministral-3-8B-Instruct-2512 | 54.4 | #1858 |
| 12 | MedGemma-4B-IT | 44.08 | #2716 |
| 13 | Gemma 2 9B (IT) | 43.87 | #2262 |
| 14 | Gemma 2 2B (IT) | 34.97 | #6876 |
| 15 | DeepSeek-R1-Distill-Qwen-7B | 34.02 | #7996 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=surgiq · How It Works · Data refreshed daily, snapshot 2026-11-13.