KorMedMCQA-V
Models answer Korean medical licensing questions using clinical images. The full test contains 1,534 questions and 2,043 images from 2012 through 2023.
The authors publish accuracy for 51 vision language models. The shorter 2022 and 2023 test split contains 304 questions.
- Made by
- Byungjin Choi, Seongsu Bae, Sunjun Kweon, and Edward Choi
- Released
- February 2026
- Kind
- Medicine
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Subject
- Clinical medicine
- Language
- Korean
- Interaction
- Single turn
- Items from
- Exams and contests
- Refresh
- One fixed set
- Run by
- Anyone
- Based on
- KorMedMCQA
- Occupations
- Healthcare practitioners and technical
- In views
- Vision, Biology and medicine
- Strongest model tested
- GPT-5.2 #51 on the capability index
- Last model added
- December 2025
- Scores refreshed
- November 20th, 2026
2 models share the top score of 96.9% among models released by December 2025 2 models share the top score of 96.9% among models released by December 2025
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in December 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 96.9
Gemini 3 Pro
- 96.9
Gemini 3 Flash
- 93.9 GPT-5
- 93.9 GPT-5.2
- 90.1 GPT-5 mini
- 83.7
Qwen3-VL-32B-Thinking
- 80.4
Qwen3-VL-30B-A3B-Thinking
- 78.7 GLM-4.6V
- 76.5
Qwen3-VL-32B-Instruct
- 74.2
Qwen3-VL-8B-Thinking
- 72.7
Ministral 3 14B
- 71.7
Qwen3-VL-30B-A3B-Instruct
- 67.2
InternVL3.5-38B
- 66.4
InternVL3.5-30B-A3B
- 66.1 Lingshu-32B
- 65.9
Qwen3-VL-4B-Thinking
- 65.5 GLM-4.6V-Flash
- 65.1 Ministral-3-8B-Reasoning-2512
- 64.7 Hulu-Med-32B
- 63.0
Qwen2.5-VL-32B
- 57.3
Gemma 3 27B
- 56.8 MedGemma-27B-IT
- 56.5 Ministral-3-14B-Instruct-2512
- 55.1 Hulu-Med-14B
- 54.3
Gemma 3 12B
- 53.8
Qwen3-VL-8B-Instruct
- 53.8
Ministral 3 8B Instruct
- 53.6
InternVL3.5-14B
- 47.4
InternVL3.5-8B
- 46.6 Lingshu-7B
- 46.6
Qwen2.5-VL-7B
- 46.1 Hulu-Med-4B
- 45.7
Qwen3-VL-4B-Instruct
- 45.2
Ministral 3 3B Reasoning
- 44.4 MedGemma-1.5-4B-IT
- 43.9 Hulu-Med-7B
- 43.3
InternVL3.5-4B
- 43.2 VARCO-VISION-2.0-14B
- 41.8 A.X-4.0-VL-Light
- 41.8
Ministral 3 3B Instruct
- 40.4
Qwen3-VL-2B-Thinking
- 37.6 Kimi-VL-A3B-Thinking
- 34.4 MedGemma-4B-IT
- 34.3
Gemma 3 4B
- 31.3
Qwen3-VL-2B-Instruct
- 30.8 Kimi-VL-A3B-Instruct
- 30.7 Kanana-1.5-V-3B-Instruct
- 29.8 InternVL3.5-2B
- 26.3 HyperCLOVAX-SEED-Vision-Instruct-3B
- 24.4 VARCO-VISION-2.0-1.7B
- 22.4 InternVL3.5-1B
- 1
Gemini 3 Pro Reasoning 96.9%
- 2
Gemini 3 Flash Reasoning 96.9%
- 3 GPT-5 Reasoning 93.9%
- 4 GPT-5.2 Reasoning 93.9%
- 5 GPT-5 mini Reasoning 90.1%
- 6
Qwen3-VL-32B-Thinking Reasoning 83.7%
- 7
Qwen3-VL-30B-A3B-Thinking Reasoning 80.4%
- 8 GLM-4.6V Reasoning 78.7%
- 9
Qwen3-VL-32B-Instruct 76.5%
- 10
Qwen3-VL-8B-Thinking Reasoning 74.2%
- 11
Ministral 3 14B Reasoning 72.7%
- 12
Qwen3-VL-30B-A3B-Instruct 71.7%
- 13
InternVL3.5-38B 67.2%
- 14
InternVL3.5-30B-A3B 66.4%
- 15 Lingshu-32B 66.1%
- 16
Qwen3-VL-4B-Thinking Reasoning 65.9%
- 17 GLM-4.6V-Flash Reasoning 65.5%
- 18 Ministral-3-8B-Reasoning-2512 Reasoning 65.1%
- 19 Hulu-Med-32B 64.7%
- 20
Qwen2.5-VL-32B 63%
- 21
Gemma 3 27B 57.3%
- 22 MedGemma-27B-IT 56.8%
- 23 Ministral-3-14B-Instruct-2512 56.5%
- 24 Hulu-Med-14B 55.1%
- 25
Gemma 3 12B 54.3%
- 26
Qwen3-VL-8B-Instruct 53.8%
- 27
Ministral 3 8B Instruct 53.8%
- 28
InternVL3.5-14B 53.6%
- 29
InternVL3.5-8B 47.4%
- 30 Lingshu-7B 46.6%
- 31
Qwen2.5-VL-7B 46.6%
- 32 Hulu-Med-4B 46.1%
- 33
Qwen3-VL-4B-Instruct 45.7%
- 34
Ministral 3 3B Reasoning Reasoning 45.2%
- 35 MedGemma-1.5-4B-IT 44.4%
- 36 Hulu-Med-7B 43.9%
- 37
InternVL3.5-4B 43.3%
- 38 VARCO-VISION-2.0-14B 43.2%
- 39 A.X-4.0-VL-Light 41.8%
- 40
Ministral 3 3B Instruct 41.8%
- 41
Qwen3-VL-2B-Thinking Reasoning 40.4%
- 42 Kimi-VL-A3B-Thinking Reasoning 37.6%
- 43 MedGemma-4B-IT 34.4%
- 44
Gemma 3 4B 34.3%
- 45
Qwen3-VL-2B-Instruct 31.3%
- 46 Kimi-VL-A3B-Instruct 30.8%
- 47 Kanana-1.5-V-3B-Instruct 30.7%
- 48 InternVL3.5-2B 29.8%
- 49 HyperCLOVAX-SEED-Vision-Instruct-3B 26.3%
- 50 VARCO-VISION-2.0-1.7B 24.4%
- 51 InternVL3.5-1B 22.4%
Source: Official KorMedMCQA-V leaderboard
- 96.9
Gemini 3 Pro
- 96.9
Gemini 3 Flash
- 93.9 GPT-5
- 93.9 GPT-5.2
- 90.1 GPT-5 mini
- 83.7
Qwen3-VL-32B-Thinking
- 80.4
Qwen3-VL-30B-A3B-Thinking
- 78.7 GLM-4.6V
- 76.5
Qwen3-VL-32B-Instruct
- 74.2
Qwen3-VL-8B-Thinking
- 72.7
Ministral 3 14B
- 71.7
Qwen3-VL-30B-A3B-Instruct
- 67.2
InternVL3.5-38B
- 66.4
InternVL3.5-30B-A3B
- 65.9
Qwen3-VL-4B-Thinking
- 65.5 GLM-4.6V-Flash
- 63.0
Qwen2.5-VL-32B
- 57.3
Gemma 3 27B
- 54.3
Gemma 3 12B
- 53.8
Qwen3-VL-8B-Instruct
- 53.8
Ministral 3 8B Instruct
- 53.6
InternVL3.5-14B
- 47.4
InternVL3.5-8B
- 46.6
Qwen2.5-VL-7B
- 45.7
Qwen3-VL-4B-Instruct
- 45.2
Ministral 3 3B Reasoning
- 43.3
InternVL3.5-4B
- 41.8
Ministral 3 3B Instruct
- 40.4
Qwen3-VL-2B-Thinking
- 37.6 Kimi-VL-A3B-Thinking
- 34.3
Gemma 3 4B
- 31.3
Qwen3-VL-2B-Instruct
- 30.8 Kimi-VL-A3B-Instruct
- 1
Gemini 3 Pro Reasoning 96.9%
- 2
Gemini 3 Flash Reasoning 96.9%
- 3 GPT-5 Reasoning 93.9%
- 4 GPT-5.2 Reasoning 93.9%
- 5 GPT-5 mini Reasoning 90.1%
- 6
Qwen3-VL-32B-Thinking Reasoning 83.7%
- 7
Qwen3-VL-30B-A3B-Thinking Reasoning 80.4%
- 8 GLM-4.6V Reasoning 78.7%
- 9
Qwen3-VL-32B-Instruct 76.5%
- 10
Qwen3-VL-8B-Thinking Reasoning 74.2%
- 11
Ministral 3 14B Reasoning 72.7%
- 12
Qwen3-VL-30B-A3B-Instruct 71.7%
- 13
InternVL3.5-38B 67.2%
- 14
InternVL3.5-30B-A3B 66.4%
- 15
Qwen3-VL-4B-Thinking Reasoning 65.9%
- 16 GLM-4.6V-Flash Reasoning 65.5%
- 17
Qwen2.5-VL-32B 63%
- 18
Gemma 3 27B 57.3%
- 19
Gemma 3 12B 54.3%
- 20
Qwen3-VL-8B-Instruct 53.8%
- 21
Ministral 3 8B Instruct 53.8%
- 22
InternVL3.5-14B 53.6%
- 23
InternVL3.5-8B 47.4%
- 24
Qwen2.5-VL-7B 46.6%
- 25
Qwen3-VL-4B-Instruct 45.7%
- 26
Ministral 3 3B Reasoning Reasoning 45.2%
- 27
InternVL3.5-4B 43.3%
- 28
Ministral 3 3B Instruct 41.8%
- 29
Qwen3-VL-2B-Thinking Reasoning 40.4%
- 30 Kimi-VL-A3B-Thinking Reasoning 37.6%
- 31
Gemma 3 4B 34.3%
- 32
Qwen3-VL-2B-Instruct 31.3%
- 33 Kimi-VL-A3B-Instruct 30.8%
Source: Official KorMedMCQA-V leaderboard
| # | Entry | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | | 96.9% | Reasoning | Nov 2025 | Official leaderboard | |
| 2 | | 96.9% | Reasoning | Dec 2025 | Official leaderboard | |
| 3 | GPT-5 | 93.9% | Reasoning | Aug 2025 | Official leaderboard | |
| 4 | GPT-5.2 | 93.9% | Reasoning | Dec 2025 | Official leaderboard | |
| 5 | GPT-5 mini | 90.1% | Reasoning | Aug 2025 | Official leaderboard | |
| 6 | | 83.7% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 7 | | 80.4% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 8 | GLM-4.6V | 78.7% | Reasoning | Dec 2025 | huggingface.co | Official leaderboard |
| 9 | | 76.5% | Oct 2025 | huggingface.co | Official leaderboard | |
| 10 | | 74.2% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 11 | | 72.7% | Reasoning | Dec 2025 | huggingface.co | Official leaderboard |
| 12 | | 71.7% | Oct 2025 | huggingface.co | Official leaderboard | |
| 13 | | 67.2% | Aug 2025 | huggingface.co | Official leaderboard | |
| 14 | | 66.4% | Aug 2025 | huggingface.co | Official leaderboard | |
| 15 | Lingshu-32B 32B | 66.1% | huggingface.co | Official leaderboard | ||
| 16 | | 65.9% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 17 | GLM-4.6V-Flash | 65.5% | Reasoning | Dec 2025 | huggingface.co | Official leaderboard |
| 18 | Ministral-3-8B-Reasoning-2512 8B | 65.1% | Reasoning | huggingface.co | Official leaderboard | |
| 19 | Hulu-Med-32B 32B | 64.7% | huggingface.co | Official leaderboard | ||
| 20 | | 63% | Mar 2025 | huggingface.co | Official leaderboard | |
| 21 | | 57.3% | Mar 2025 | huggingface.co | Official leaderboard | |
| 22 | MedGemma-27B-IT 27B | 56.8% | huggingface.co | Official leaderboard | ||
| 23 | Ministral-3-14B-Instruct-2512 14B | 56.5% | huggingface.co | Official leaderboard | ||
| 24 | Hulu-Med-14B 14B | 55.1% | huggingface.co | Official leaderboard | ||
| 25 | | 54.3% | Mar 2025 | huggingface.co | Official leaderboard | |
| 26 | | 53.8% | Oct 2025 | huggingface.co | Official leaderboard | |
| 27 | | 53.8% | Dec 2025 | huggingface.co | Official leaderboard | |
| 28 | | 53.6% | Aug 2025 | huggingface.co | Official leaderboard | |
| 29 | | 47.4% | Aug 2025 | huggingface.co | Official leaderboard | |
| 30 | Lingshu-7B 7B | 46.6% | huggingface.co | Official leaderboard | ||
| 31 | | 46.6% | Jan 2025 | huggingface.co | Official leaderboard | |
| 32 | Hulu-Med-4B 4B | 46.1% | huggingface.co | Official leaderboard | ||
| 33 | | 45.7% | Oct 2025 | huggingface.co | Official leaderboard | |
| 34 | | 45.2% | Reasoning | Dec 2025 | huggingface.co | Official leaderboard |
| 35 | MedGemma-1.5-4B-IT 4B | 44.4% | huggingface.co | Official leaderboard | ||
| 36 | Hulu-Med-7B 7B | 43.9% | huggingface.co | Official leaderboard | ||
| 37 | | 43.3% | Aug 2025 | huggingface.co | Official leaderboard | |
| 38 | VARCO-VISION-2.0-14B 14B | 43.2% | huggingface.co | Official leaderboard | ||
| 39 | A.X-4.0-VL-Light 7B | 41.8% | huggingface.co | Official leaderboard | ||
| 40 | | 41.8% | Dec 2025 | huggingface.co | Official leaderboard | |
| 41 | | 40.4% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 42 | Kimi-VL-A3B-Thinking | 37.6% | Reasoning | Apr 2025 | huggingface.co | Official leaderboard |
| 43 | MedGemma-4B-IT 4B | 34.4% | huggingface.co | Official leaderboard | ||
| 44 | | 34.3% | Mar 2025 | huggingface.co | Official leaderboard | |
| 45 | | 31.3% | Oct 2025 | huggingface.co | Official leaderboard | |
| 46 | Kimi-VL-A3B-Instruct | 30.8% | Apr 2025 | huggingface.co | Official leaderboard | |
| 47 | Kanana-1.5-V-3B-Instruct 3B | 30.7% | huggingface.co | Official leaderboard | ||
| 48 | InternVL3.5-2B 2B | 29.8% | huggingface.co | Official leaderboard | ||
| 49 | HyperCLOVAX-SEED-Vision-Instruct-3B 3B | 26.3% | huggingface.co | Official leaderboard | ||
| 50 | VARCO-VISION-2.0-1.7B 1.7B | 24.4% | huggingface.co | Official leaderboard | ||
| 51 | InternVL3.5-1B 1B | 22.4% | huggingface.co | Official leaderboard |
| # | Model | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | | 96.9% | Reasoning | Nov 2025 | Official leaderboard | |
| 2 | | 96.9% | Reasoning | Dec 2025 | Official leaderboard | |
| 3 | GPT-5 | 93.9% | Reasoning | Aug 2025 | Official leaderboard | |
| 4 | GPT-5.2 | 93.9% | Reasoning | Dec 2025 | Official leaderboard | |
| 5 | GPT-5 mini | 90.1% | Reasoning | Aug 2025 | Official leaderboard | |
| 6 | | 83.7% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 7 | | 80.4% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 8 | GLM-4.6V | 78.7% | Reasoning | Dec 2025 | huggingface.co | Official leaderboard |
| 9 | | 76.5% | Oct 2025 | huggingface.co | Official leaderboard | |
| 10 | | 74.2% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 11 | | 72.7% | Reasoning | Dec 2025 | huggingface.co | Official leaderboard |
| 12 | | 71.7% | Oct 2025 | huggingface.co | Official leaderboard | |
| 13 | | 67.2% | Aug 2025 | huggingface.co | Official leaderboard | |
| 14 | | 66.4% | Aug 2025 | huggingface.co | Official leaderboard | |
| 15 | | 65.9% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 16 | GLM-4.6V-Flash | 65.5% | Reasoning | Dec 2025 | huggingface.co | Official leaderboard |
| 17 | | 63% | Mar 2025 | huggingface.co | Official leaderboard | |
| 18 | | 57.3% | Mar 2025 | huggingface.co | Official leaderboard | |
| 19 | | 54.3% | Mar 2025 | huggingface.co | Official leaderboard | |
| 20 | | 53.8% | Oct 2025 | huggingface.co | Official leaderboard | |
| 21 | | 53.8% | Dec 2025 | huggingface.co | Official leaderboard | |
| 22 | | 53.6% | Aug 2025 | huggingface.co | Official leaderboard | |
| 23 | | 47.4% | Aug 2025 | huggingface.co | Official leaderboard | |
| 24 | | 46.6% | Jan 2025 | huggingface.co | Official leaderboard | |
| 25 | | 45.7% | Oct 2025 | huggingface.co | Official leaderboard | |
| 26 | | 45.2% | Reasoning | Dec 2025 | huggingface.co | Official leaderboard |
| 27 | | 43.3% | Aug 2025 | huggingface.co | Official leaderboard | |
| 28 | | 41.8% | Dec 2025 | huggingface.co | Official leaderboard | |
| 29 | | 40.4% | Reasoning | Oct 2025 | huggingface.co | Official leaderboard |
| 30 | Kimi-VL-A3B-Thinking | 37.6% | Reasoning | Apr 2025 | huggingface.co | Official leaderboard |
| 31 | | 34.3% | Mar 2025 | huggingface.co | Official leaderboard | |
| 32 | | 31.3% | Oct 2025 | huggingface.co | Official leaderboard | |
| 33 | Kimi-VL-A3B-Instruct | 30.8% | Apr 2025 | huggingface.co | Official leaderboard |
Rankings by part
We read 12 parts from the benchmark’s own leaderboard in 2 groups: modality and image count. Each ranks models on one part of the test. Gemini 3 Pro leads 5 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
xray
33 models we track have a score here.
- 1
Gemini 3 Pro 97%
- 2
Gemini 3 Flash 96.2%
- 3 GPT-5.2 ↑1 94.2%
- 4 GPT-5 ↓1 93.6%
- 5 GPT-5 mini 89.1%
- 6
Qwen3-VL-32B-Thinking 83.9%
- 7
Qwen3-VL-30B-A3B-Thinking 80.7%
- 8 GLM-4.6V 78.4%
- 9
Qwen3-VL-32B-Instruct 75.9%
- 10
Qwen3-VL-8B-Thinking 75.1%
ct
33 models we track have a score here.
- 1
Gemini 3 Pro 97.9%
- 2
Gemini 3 Flash 97.9%
- 3 GPT-5.2 ↑1 89.6%
- 4 GPT-5 ↓1 89.1%
- 5 GPT-5 mini 85.4%
- 6
Qwen3-VL-32B-Thinking 78%
- 7 GLM-4.6V ↑1 75%
- 8
Qwen3-VL-30B-A3B-Thinking ↓1 73.8%
- 9
Qwen3-VL-32B-Instruct 71.9%
- 10
Ministral 3 14B ↑1 70.3%
ecg
33 models we track have a score here.
- 1
Gemini 3 Pro 95.9%
- 2
Gemini 3 Flash 93.9%
- 3 GPT-5.2 ↑1 92.9%
- 4 GPT-5 ↓1 89.8%
- 5
Qwen3-VL-32B-Thinking ↑1 86.4%
- 6 GPT-5 mini ↓1 85.7%
- 7
Qwen3-VL-32B-Instruct ↑2 83.3%
- 8
Qwen3-VL-30B-A3B-Thinking ↓1 82.3%
- 9 GLM-4.6V ↓1 80.3%
- 10
Qwen3-VL-30B-A3B-Instruct ↑2 75.5%
us
33 models we track have a score here.
- 1 GPT-5.2 ↑3 98.1%
- 2 GPT-5 ↑1 97.2%
- 3
Gemini 3 Pro ↓2 97.2%
- 4
Gemini 3 Flash ↓2 96.2%
- 5 GPT-5 mini 90.6%
- 6
Qwen3-VL-30B-A3B-Thinking ↑1 81.1%
- 7
Qwen3-VL-32B-Thinking ↓1 81.1%
- 8
Qwen3-VL-32B-Instruct ↑1 77%
- 9
Qwen3-VL-8B-Thinking ↑1 76.7%
- 10
Ministral 3 14B ↑1 73.6%
endoscopy
33 models we track have a score here.
- 1
Gemini 3 Flash ↑1 94.6%
- 2
Gemini 3 Pro ↓1 93.5%
- 3 GPT-5 92.4%
- 4 GPT-5.2 90.2%
- 5 GPT-5 mini 88%
- 6
Qwen3-VL-32B-Thinking 76.4%
- 7
Qwen3-VL-30B-A3B-Thinking 75.4%
- 8
Ministral 3 14B ↑3 73.9%
- 9 GLM-4.6V ↓1 72.8%
- 10
Qwen3-VL-8B-Thinking 69.6%
nst
33 models we track have a score here.
- 1
Gemini 3 Flash ↑1 93.5%
- 2
Gemini 3 Pro ↓1 91.3%
- 3 GPT-5 mini ↑2 89.1%
- 4 GPT-5 ↓1 87%
- 5 GPT-5.2 ↓1 84.8%
- 6
Qwen3-VL-32B-Thinking 71.7%
- 7
Qwen3-VL-30B-A3B-Thinking 70.3%
- 8
Qwen3-VL-8B-Thinking ↑2 68.8%
- 9
Qwen3-VL-32B-Instruct 64.5%
- 10
Qwen3-VL-30B-A3B-Instruct ↑2 63%
pbs
33 models we track have a score here.
- 1 GPT-5 ↑2 100%
- 2
Gemini 3 Pro ↓1 100%
- 3 GPT-5.2 ↑1 100%
- 4
Gemini 3 Flash ↓2 100%
- 5 GPT-5 mini 97.7%
- 6
Qwen3-VL-32B-Thinking 95.3%
- 7
Qwen3-VL-30B-A3B-Thinking 90.7%
- 8 GLM-4.6V 86.8%
- 9
Qwen3-VL-32B-Instruct 86%
- 10
Qwen3-VL-8B-Thinking 84.5%
mri
33 models we track have a score here.
- 1 GPT-5 ↑2 100%
- 2
Gemini 3 Pro ↓1 100%
- 3
Gemini 3 Flash ↓1 100%
- 4 GPT-5 mini ↑1 96.3%
- 5 GPT-5.2 ↓1 96.3%
- 6
Qwen3-VL-32B-Thinking 88.9%
- 7
Qwen3-VL-30B-A3B-Thinking 87.7%
- 8
Ministral 3 14B ↑3 86.4%
- 9 GLM-4.6V ↓1 86.4%
- 10
Qwen3-VL-32B-Instruct ↓1 85.2%
other
33 models we track have a score here.
- 1
Gemini 3 Flash ↑1 98.3%
- 2
Gemini 3 Pro ↓1 97.4%
- 3 GPT-5 96.3%
- 4 GPT-5.2 95.5%
- 5 GPT-5 mini 93.3%
- 6
Qwen3-VL-32B-Thinking 87%
- 7 GLM-4.6V ↑1 83.3%
- 8
Qwen3-VL-30B-A3B-Thinking ↓1 83%
- 9
Qwen3-VL-32B-Instruct 79.1%
- 10
Qwen3-VL-8B-Thinking 75.8%
1
33 models we track have a score here.
- 1
Gemini 3 Pro 97.7%
- 2
Gemini 3 Flash 97.6%
- 3 GPT-5 94.7%
- 4 GPT-5.2 94.7%
- 5 GPT-5 mini 91.3%
- 6
Qwen3-VL-32B-Thinking 85.2%
- 7
Qwen3-VL-30B-A3B-Thinking 81.7%
- 8 GLM-4.6V 80%
- 9
Qwen3-VL-32B-Instruct 77.9%
- 10
Qwen3-VL-8B-Thinking 75.7%
2
33 models we track have a score here.
- 1
Gemini 3 Pro 96%
- 2
Gemini 3 Flash 95.5%
- 3 GPT-5.2 ↑1 92.7%
- 4 GPT-5 ↓1 92%
- 5 GPT-5 mini 87.3%
- 6
Qwen3-VL-32B-Thinking 80.5%
- 7
Qwen3-VL-30B-A3B-Thinking 78.3%
- 8 GLM-4.6V 75.1%
- 9
Qwen3-VL-32B-Instruct 74.5%
- 10
Qwen3-VL-8B-Thinking 71.4%
3+
33 models we track have a score here.
- 1 GPT-5 ↑2 92.3%
- 2
Gemini 3 Flash 92.3%
- 3 GPT-5 mini ↑2 87.2%
- 4
Gemini 3 Pro ↓3 87.2%
- 5 GPT-5.2 ↓1 84.6%
- 6 GLM-4.6V ↑2 80.3%
- 7
Qwen3-VL-32B-Thinking ↓1 76.9%
- 8
Ministral 3 14B ↑3 69.2%
- 9
Qwen3-VL-30B-A3B-Thinking ↓2 67.5%
- 10
Qwen3-VL-8B-Thinking 65.8%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official KorMedMCQA-V leaderboard
GPT-5 on the official leaderboard passed 85% in August 2025
Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official KorMedMCQA-V leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Law, finance and medicine
- MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy.
Qwen 3.8 Max 80.4%
- HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
- HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
- FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy.
Ling-3.1-flash 81.7%
- KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
- Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%
Data audit
All 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (51 entries).
- Pass Models scored 33 models we track, plus 18 other entries on its leaderboard.
- Note Newest score The newest came 11 months ago, in December 2025, for Gemini 3 Flash. That’s expected for a saturated test.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-5.2 (#51).
- Note Headroom Saturated. The best is 96.9%, past 85% since August 2025.
Leaderboard import 3 pass
- Pass Importer We read the board on kormedmcqa-v.github.io: 33 rows for models we track and 18 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.