KMMLU-Pro 2026
Version 2026
Models answer questions from the 2026 Korean national professional licensing exams. The dataset uses the KMMLU-Pro evaluation format.
The dataset also includes a nursing exam. The authors explain how to exclude nursing for comparisons with the original 14 licenses.
- Made by
- HAERAE-HUB
- Released
- October 2026
- Kind
- Knowledge in many languages
- Level
- University
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Subject
- Law, Clinical medicine
- Language
- Korean
- Interaction
- Single turn
- Items from
- Exams and contests
- Refresh
- New versions replace it
- Run by
- Anyone
- Based on
- KMMLU-Pro
- Occupations
- Legal, Healthcare practitioners and technical
- In views
- Biology and medicine, Law and finance
- Strongest model tested
- o3 #84 on the capability index
- Last model added
- July 2025
- Scores refreshed
- November 20th, 2026
o1 leads with 78.1% among models released by July 2025 o1 leads with 78.1% among models released by July 2025
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. 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 July 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
- 78.1 o1
- 77.7 Claude 3.7 Sonnet
- 73.6 o3
- 73.0 GPT-4.1 (2025-04-14)
- 71.3 DeepSeek R1 (671B)
- 69.7 o4-mini (2025-04-16)
- 68.4 Grok 3
- 68.2
Qwen3-235B-A22B
- 68.1
Llama 4 Maverick
- 67.7
EXAONE 4.0 32B
- 65.1 Grok-3 mini
- 63.9
QwQ-32B
- 62.2 GPT-4.1 mini (2025-04-14)
- 62.0 o3-mini
- 61.1
Qwen3-32B
- 60.8 DeepSeek V3 (671B)
- 60.5
Qwen3-30B-A3B
- 60.2
Qwen3-14B
- 58.1
Llama 4 Scout
- 57.5 C4AI Command A (111B)
- 53.2
Llama 3.3 70B Instruct
- 51.0
Gemma 3 27B
- 49.5 Mistral Small 3.1 Instruct (24B)
- 45.8
Gemma 3 12B
- 45.3 Phi-4 (14B)
- 31.3
Aya Expanse 32B
- 1 o1 78.1%
- 2 Claude 3.7 Sonnet w/ thinking 77.7%
- 3 o3 73.6%
- 4 GPT-4.1 (2025-04-14) 73%
- 5 DeepSeek R1 (671B) 71.3%
- 6 o4-mini (2025-04-16) 69.7%
- 7 Grok 3 68.4%
- 8
Qwen3-235B-A22B w/ thinking 68.2%
- 9
Llama 4 Maverick 68.1%
- 10
EXAONE 4.0 32B w/ thinking 67.7%
- 11 Grok-3 mini 65.1%
- 12
QwQ-32B 63.9%
- 13 GPT-4.1 mini (2025-04-14) 62.2%
- 14 o3-mini 62%
- 15
Qwen3-32B w/ thinking 61.1%
- 16 DeepSeek V3 (671B) 60.8%
- 17
Qwen3-30B-A3B w/ thinking 60.5%
- 18
Qwen3-14B w/ thinking 60.2%
- 19
Llama 4 Scout 58.1%
- 20 C4AI Command A (111B) 57.5%
- 21
Llama 3.3 70B Instruct 53.2%
- 22
Gemma 3 27B 51%
- 23 Mistral Small 3.1 Instruct (24B) 49.5%
- 24
Gemma 3 12B 45.8%
- 25 Phi-4 (14B) 45.3%
- 26
Aya Expanse 32B 31.3%
Source: The KMMLU-Pro 2026 paper
- 78.1 o1
- 77.7 Claude 3.7 Sonnet
- 73.6 o3
- 68.4 Grok 3
- 68.2
Qwen3-235B-A22B
- 68.1
Llama 4 Maverick
- 67.7
EXAONE 4.0 32B
- 65.1 Grok-3 mini
- 63.9
QwQ-32B
- 62.0 o3-mini
- 61.1
Qwen3-32B
- 60.5
Qwen3-30B-A3B
- 60.2
Qwen3-14B
- 58.1
Llama 4 Scout
- 53.2
Llama 3.3 70B Instruct
- 51.0
Gemma 3 27B
- 45.8
Gemma 3 12B
- 31.3
Aya Expanse 32B
- 1 o1 78.1%
- 2 Claude 3.7 Sonnet w/ thinking 77.7%
- 3 o3 73.6%
- 4 Grok 3 68.4%
- 5
Qwen3-235B-A22B w/ thinking 68.2%
- 6
Llama 4 Maverick 68.1%
- 7
EXAONE 4.0 32B w/ thinking 67.7%
- 8 Grok-3 mini 65.1%
- 9
QwQ-32B 63.9%
- 10 o3-mini 62%
- 11
Qwen3-32B w/ thinking 61.1%
- 12
Qwen3-30B-A3B w/ thinking 60.5%
- 13
Qwen3-14B w/ thinking 60.2%
- 14
Llama 4 Scout 58.1%
- 15
Llama 3.3 70B Instruct 53.2%
- 16
Gemma 3 27B 51%
- 17
Gemma 3 12B 45.8%
- 18
Aya Expanse 32B 31.3%
Source: The KMMLU-Pro 2026 paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | o1 | 78.1% | Dec 2024 | Official leaderboard | |
| 2 | Claude 3.7 Sonnet | 77.7% | w/ thinking | Feb 2025 | Official leaderboard |
| 3 | o3 | 73.6% | Apr 2025 | Official leaderboard | |
| 4 | GPT-4.1 (2025-04-14) | 73% | Official leaderboard | ||
| 5 | DeepSeek R1 (671B) | 71.3% | Official leaderboard | ||
| 6 | o4-mini (2025-04-16) | 69.7% | Official leaderboard | ||
| 7 | Grok 3 | 68.4% | Apr 2025 | Official leaderboard | |
| 8 | | 68.2% | w/ thinking | Apr 2025 | Official leaderboard |
| 9 | | 68.1% | Apr 2025 | Official leaderboard | |
| 10 | | 67.7% | w/ thinking | Jul 2025 | Official leaderboard |
| 11 | Grok-3 mini | 65.1% | Apr 2025 | Official leaderboard | |
| 12 | | 63.9% | Mar 2025 | Official leaderboard | |
| 13 | GPT-4.1 mini (2025-04-14) | 62.2% | Official leaderboard | ||
| 14 | o3-mini | 62% | Jan 2025 | Official leaderboard | |
| 15 | | 61.1% | w/ thinking | Apr 2025 | Official leaderboard |
| 16 | DeepSeek V3 (671B) | 60.8% | Official leaderboard | ||
| 17 | | 60.5% | w/ thinking | Apr 2025 | Official leaderboard |
| 18 | | 60.2% | w/ thinking | Apr 2025 | Official leaderboard |
| 19 | | 58.1% | Apr 2025 | Official leaderboard | |
| 20 | C4AI Command A (111B) | 57.5% | Official leaderboard | ||
| 21 | | 53.2% | Dec 2024 | Official leaderboard | |
| 22 | | 51% | Mar 2025 | Official leaderboard | |
| 23 | Mistral Small 3.1 Instruct (24B) | 49.5% | Official leaderboard | ||
| 24 | | 45.8% | Mar 2025 | Official leaderboard | |
| 25 | Phi-4 (14B) | 45.3% | Official leaderboard | ||
| 26 | | 31.3% | Oct 2024 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | o1 | 78.1% | Dec 2024 | Official leaderboard | |
| 2 | Claude 3.7 Sonnet | 77.7% | w/ thinking | Feb 2025 | Official leaderboard |
| 3 | o3 | 73.6% | Apr 2025 | Official leaderboard | |
| 4 | Grok 3 | 68.4% | Apr 2025 | Official leaderboard | |
| 5 | | 68.2% | w/ thinking | Apr 2025 | Official leaderboard |
| 6 | | 68.1% | Apr 2025 | Official leaderboard | |
| 7 | | 67.7% | w/ thinking | Jul 2025 | Official leaderboard |
| 8 | Grok-3 mini | 65.1% | Apr 2025 | Official leaderboard | |
| 9 | | 63.9% | Mar 2025 | Official leaderboard | |
| 10 | o3-mini | 62% | Jan 2025 | Official leaderboard | |
| 11 | | 61.1% | w/ thinking | Apr 2025 | Official leaderboard |
| 12 | | 60.5% | w/ thinking | Apr 2025 | Official leaderboard |
| 13 | | 60.2% | w/ thinking | Apr 2025 | Official leaderboard |
| 14 | | 58.1% | Apr 2025 | Official leaderboard | |
| 15 | | 53.2% | Dec 2024 | Official leaderboard | |
| 16 | | 51% | Mar 2025 | Official leaderboard | |
| 17 | | 45.8% | Mar 2025 | Official leaderboard | |
| 18 | | 31.3% | Oct 2024 | Official leaderboard |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The KMMLU-Pro 2026 paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| o1 | 78.1% | ||||
| Claude 3.7 Sonnet | w/ thinking | 77.7% | |||
| Claude 3.7 Sonnet | 74.5% | ||||
| o3 | 73.6% | ||||
| Grok 3 | 68.4% | ||||
| w/ thinking | 68.2% | ||||
| 68.1% | |||||
| w/ thinking | 67.7% | ||||
| Grok-3 mini | 65.1% | ||||
| 63.9% | |||||
| 62.1% | |||||
| o3-mini | 62% | ||||
| w/ thinking | 61.1% | ||||
| w/ thinking | 60.5% | ||||
| w/ thinking | 60.2% | ||||
| 60% | |||||
| 58.9% | |||||
| 58.1% | |||||
| 53.2% | |||||
| 53% | |||||
| 52.3% | |||||
| 51% | |||||
| 45.8% | |||||
| 31.3% |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The KMMLU-Pro 2026 paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| o1 | 78.1% | ||||
| Claude 3.7 Sonnet | w/ thinking | 77.7% | |||
| Claude 3.7 Sonnet | 74.5% | ||||
| o3 | 73.6% | ||||
| Grok 3 | 68.4% | ||||
| w/ thinking | 68.2% | ||||
| 68.1% | |||||
| w/ thinking | 67.7% | ||||
| Grok-3 mini | 65.1% | ||||
| 63.9% | |||||
| 62.1% | |||||
| o3-mini | 62% | ||||
| w/ thinking | 61.1% | ||||
| w/ thinking | 60.5% | ||||
| w/ thinking | 60.2% | ||||
| 60% | |||||
| 58.9% | |||||
| 58.1% | |||||
| 53.2% | |||||
| 53% | |||||
| 52.3% | |||||
| 51% | |||||
| 45.8% | |||||
| 31.3% |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The KMMLU-Pro 2026 paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official KMMLU-Pro 2026 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.
Versions
More in Multilingual
- MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
- WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET.
K-EXAONE 90.5%
- MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language.
Qwen3.7-Max 87%
- CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories.
Solar Open 2 90.7%
- Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
- MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 16 months ago, in July 2025, for EXAONE 4.0 32B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (26 entries).
- Pass Models scored 18 models we track, plus 8 other entries on its leaderboard.
- Warning Newest score The newest came 16 months ago, in July 2025, for EXAONE 4.0 32B.
- Note Frontier models None of the top 10 has a score. The strongest tested is o3 (#84).
- Pass Headroom The best is 78.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 24 rows for models we track and 8 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.