GreekMMLU
Models answer questions originally written in Greek for academic, professional, and government exams across 45 subjects.
EuroEval reports the Matthews correlation coefficient for its multiple-choice version.
- Made by
- Yang Zhang and colleagues
- Released
- February 2026
- Kind
- Knowledge in many languages
- Level
- University
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Language
- Greek
- Interaction
- Single turn
- Items from
- Exams and contests
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Law and finance
- Strongest model tested
- GPT-5.2 #51 on the capability index
- Last model added
- December 2025
- Scores refreshed
- November 20th, 2026
Gemini 3 Flash leads with 93.2% among models released by December 2025 Gemini 3 Flash leads with 93.2% among models released by December 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 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
- 93.2
Gemini 3 Flash
- 89.2 GPT-5.2
- 87.8 GPT-4o (snapshot unspecified)
- 82.6
Gemma 3 27B PT
- 82.2
Qwen2.5-72B
- 81.5
Llama 3.3 70B Instruct
- 81.3
Gemma 3 27B
- 80.8 Qwen3-30B-Instruct
- 80.7 Llama-3.1-70B-Instruct
- 80.4
Llama 3.1-70B
- 79.9 Qwen3-30B
- 77.3
Gemma 3 12B PT
- 77.3
Qwen3-14B
- 76.3
Gemma 3 12B
- 76.0 Qwen2.5-32B-Instruct
- 75.7
Qwen2.5-32B
- 75.0 EuroLLM-22B-Instruct-2512
- 74.1 EuroLLM-22B
- 72.8
Qwen3-8B
- 71.1
Qwen2.5-14B
- 70.9 Llama-Krikri-8B-Base
- 70.8 Llama-Krikri-8B-Instruct
- 70.8 Plutus-8B-instruct
- 70.6 Qwen2.5-14B-Instruct
- 70.5
Mistral Small 3
- 70.2 GLM-4-9B
- 68.8 GLM-4-9B-Chat
- 68.5 EuroLLM-9B-Instruct
- 68.3 qwen3-4b-instruct-2507
- 67.1
Qwen3-4B
- 66.3 EuroLLM-9B
- 65.6
Aya Expanse 8B
- 64.9 Llama-3-8B-Instruct
- 64.8
Gemma 3 4B
- 64.7 Llama-3.1-8B-Instruct
- 64.5
Gemma 3 4B PT
- 64.2
Llama 3-8B
- 64.2 Qwen2.5-7B-Instruct
- 63.7 Meltemi-7B-Instruct-v1.5
- 63.3
Llama 3.1-8B
- 63.0 Meltemi-7B-v1.5
- 63.0
Qwen2.5-7B
- 58.6 Aya-101
- 58.4 Meltemi-7B-v1
- 58.0
Qwen3-1.7B
- 56.6 mT0-xxl
- 53.4 Qwen2.5-3B-Instruct
- 51.6 Mistral-7B-Instruct-v0.3
- 50.5 Llama-3.2-3B-Instruct
- 50.0 mT0-xl
- 48.4
Llama 3.2 3B
- 47.5
Qwen2.5-3B
- 46.5 Qwen2.5-1.5B-Instruct
- 45.7
Gemma 3 1B
- 42.9
Qwen2.5-1.5B
- 41.0
Qwen3-0.6B
- 40.5 Llama-3.2-1B-Instruct
- 37.0 mT0-large
- 36.8 Llama-2-7b-chat-hf
- 36.4
Llama 3.2 1B
- 35.3
Llama 2-7B
- 35.1 BLOOMZ-1b1
- 34.9 Qwen2.5-0.5B-Instruct
- 34.7
Qwen2.5-0.5B
- 34.7 BLOOMZ-1b7
- 33.0 XGLM-4.5B
- 33.0 XGLM-1.7B
- 32.7 XGLM-2.9B
- 32.4 BLOOMZ-7b1
- 32.4 EuroLLM-1.7B-Instruct
- 32.3 EuroLLM-1.7B
- 32.2 XGLM-7.5B
- 31.9
Gemma 3 1B PT
- 1
Gemini 3 Flash zero-shot 93.2%
- 2 GPT-5.2 five-shot 89.2%
- 3 GPT-4o (snapshot unspecified) five-shot 87.8%
- 4
Gemma 3 27B PT five-shot 82.6%
- 5
Qwen2.5-72B five-shot 82.2%
- 6
Llama 3.3 70B Instruct five-shot 81.5%
- 7
Gemma 3 27B five-shot 81.3%
- 8 Qwen3-30B-Instruct five-shot 80.8%
- 9 Llama-3.1-70B-Instruct five-shot 80.7%
- 10
Llama 3.1-70B five-shot 80.4%
- 11 Qwen3-30B five-shot 79.9%
- 12
Gemma 3 12B PT five-shot 77.3%
- 13
Qwen3-14B five-shot 77.3%
- 14
Gemma 3 12B five-shot 76.3%
- 15 Qwen2.5-32B-Instruct five-shot 76%
- 16
Qwen2.5-32B five-shot 75.7%
- 17 EuroLLM-22B-Instruct-2512 five-shot 75%
- 18 EuroLLM-22B five-shot 74.1%
- 19
Qwen3-8B five-shot 72.8%
- 20
Qwen2.5-14B five-shot 71.1%
- 21 Llama-Krikri-8B-Base five-shot 70.9%
- 22 Llama-Krikri-8B-Instruct five-shot 70.8%
- 23 Plutus-8B-instruct five-shot 70.8%
- 24 Qwen2.5-14B-Instruct five-shot 70.6%
- 25
Mistral Small 3 zero-shot 70.5%
- 26 GLM-4-9B five-shot 70.2%
- 27 GLM-4-9B-Chat five-shot 68.8%
- 28 EuroLLM-9B-Instruct zero-shot 68.5%
- 29 qwen3-4b-instruct-2507 five-shot 68.3%
- 30
Qwen3-4B five-shot 67.1%
- 31 EuroLLM-9B zero-shot 66.3%
- 32
Aya Expanse 8B five-shot 65.6%
- 33 Llama-3-8B-Instruct five-shot 64.9%
- 34
Gemma 3 4B five-shot 64.8%
- 35 Llama-3.1-8B-Instruct five-shot 64.7%
- 36
Gemma 3 4B PT five-shot 64.5%
- 37
Llama 3-8B five-shot 64.2%
- 38 Qwen2.5-7B-Instruct five-shot 64.2%
- 39 Meltemi-7B-Instruct-v1.5 five-shot 63.7%
- 40
Llama 3.1-8B five-shot 63.3%
- 41 Meltemi-7B-v1.5 five-shot 63%
- 42
Qwen2.5-7B five-shot 63%
- 43 Aya-101 five-shot 58.6%
- 44 Meltemi-7B-v1 five-shot 58.4%
- 45
Qwen3-1.7B five-shot 58%
- 46 mT0-xxl zero-shot 56.6%
- 47 Qwen2.5-3B-Instruct five-shot 53.4%
- 48 Mistral-7B-Instruct-v0.3 five-shot 51.6%
- 49 Llama-3.2-3B-Instruct five-shot 50.5%
- 50 mT0-xl zero-shot 50%
- 51
Llama 3.2 3B five-shot 48.4%
- 52
Qwen2.5-3B five-shot 47.5%
- 53 Qwen2.5-1.5B-Instruct five-shot 46.5%
- 54
Gemma 3 1B five-shot 45.7%
- 55
Qwen2.5-1.5B five-shot 42.9%
- 56
Qwen3-0.6B five-shot 41%
- 57 Llama-3.2-1B-Instruct five-shot 40.5%
- 58 mT0-large zero-shot 37%
- 59 Llama-2-7b-chat-hf five-shot 36.8%
- 60
Llama 3.2 1B zero-shot 36.4%
- 61
Llama 2-7B zero-shot 35.3%
- 62 BLOOMZ-1b1 zero-shot 35.1%
- 63 Qwen2.5-0.5B-Instruct zero-shot 34.9%
- 64
Qwen2.5-0.5B zero-shot 34.7%
- 65 BLOOMZ-1b7 zero-shot 34.7%
- 66 XGLM-4.5B zero-shot 33%
- 67 XGLM-1.7B zero-shot 33%
- 68 XGLM-2.9B zero-shot 32.7%
- 69 BLOOMZ-7b1 zero-shot 32.4%
- 70 EuroLLM-1.7B-Instruct five-shot 32.4%
- 71 EuroLLM-1.7B zero-shot 32.3%
- 72 XGLM-7.5B zero-shot 32.2%
- 73
Gemma 3 1B PT zero-shot 31.9%
Source: The GreekMMLU paper
- 93.2
Gemini 3 Flash
- 89.2 GPT-5.2
- 82.6
Gemma 3 27B PT
- 82.2
Qwen2.5-72B
- 81.5
Llama 3.3 70B Instruct
- 81.3
Gemma 3 27B
- 80.4
Llama 3.1-70B
- 77.3
Gemma 3 12B PT
- 77.3
Qwen3-14B
- 76.3
Gemma 3 12B
- 75.7
Qwen2.5-32B
- 72.8
Qwen3-8B
- 71.1
Qwen2.5-14B
- 70.5
Mistral Small 3
- 70.2 GLM-4-9B
- 68.8 GLM-4-9B-Chat
- 68.3 qwen3-4b-instruct-2507
- 67.1
Qwen3-4B
- 65.6
Aya Expanse 8B
- 64.8
Gemma 3 4B
- 64.5
Gemma 3 4B PT
- 64.2
Llama 3-8B
- 63.3
Llama 3.1-8B
- 63.0
Qwen2.5-7B
- 58.0
Qwen3-1.7B
- 48.4
Llama 3.2 3B
- 47.5
Qwen2.5-3B
- 45.7
Gemma 3 1B
- 42.9
Qwen2.5-1.5B
- 41.0
Qwen3-0.6B
- 36.4
Llama 3.2 1B
- 35.3
Llama 2-7B
- 34.7
Qwen2.5-0.5B
- 31.9
Gemma 3 1B PT
- 1
Gemini 3 Flash zero-shot 93.2%
- 2 GPT-5.2 five-shot 89.2%
- 3
Gemma 3 27B PT five-shot 82.6%
- 4
Qwen2.5-72B five-shot 82.2%
- 5
Llama 3.3 70B Instruct five-shot 81.5%
- 6
Gemma 3 27B five-shot 81.3%
- 7
Llama 3.1-70B five-shot 80.4%
- 8
Gemma 3 12B PT five-shot 77.3%
- 9
Qwen3-14B five-shot 77.3%
- 10
Gemma 3 12B five-shot 76.3%
- 11
Qwen2.5-32B five-shot 75.7%
- 12
Qwen3-8B five-shot 72.8%
- 13
Qwen2.5-14B five-shot 71.1%
- 14
Mistral Small 3 zero-shot 70.5%
- 15 GLM-4-9B five-shot 70.2%
- 16 GLM-4-9B-Chat five-shot 68.8%
- 17 qwen3-4b-instruct-2507 five-shot 68.3%
- 18
Qwen3-4B five-shot 67.1%
- 19
Aya Expanse 8B five-shot 65.6%
- 20
Gemma 3 4B five-shot 64.8%
- 21
Gemma 3 4B PT five-shot 64.5%
- 22
Llama 3-8B five-shot 64.2%
- 23
Llama 3.1-8B five-shot 63.3%
- 24
Qwen2.5-7B five-shot 63%
- 25
Qwen3-1.7B five-shot 58%
- 26
Llama 3.2 3B five-shot 48.4%
- 27
Qwen2.5-3B five-shot 47.5%
- 28
Gemma 3 1B five-shot 45.7%
- 29
Qwen2.5-1.5B five-shot 42.9%
- 30
Qwen3-0.6B five-shot 41%
- 31
Llama 3.2 1B zero-shot 36.4%
- 32
Llama 2-7B zero-shot 35.3%
- 33
Qwen2.5-0.5B zero-shot 34.7%
- 34
Gemma 3 1B PT zero-shot 31.9%
Source: The GreekMMLU paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 93.2% | zero-shot | Dec 2025 | Official leaderboard |
| 2 | GPT-5.2 | 89.2% | five-shot | Dec 2025 | Official leaderboard |
| 3 | GPT-4o (snapshot unspecified) | 87.8% | five-shot | Official leaderboard | |
| 4 | | 82.6% | five-shot | Mar 2025 | Official leaderboard |
| 5 | | 82.2% | five-shot | Sep 2024 | Official leaderboard |
| 6 | | 81.5% | five-shot | Dec 2024 | Official leaderboard |
| 7 | | 81.3% | five-shot | Mar 2025 | Official leaderboard |
| 8 | Qwen3-30B-Instruct | 80.8% | five-shot | Official leaderboard | |
| 9 | Llama-3.1-70B-Instruct | 80.7% | five-shot | Official leaderboard | |
| 10 | | 80.4% | five-shot | Jul 2024 | Official leaderboard |
| 11 | Qwen3-30B | 79.9% | five-shot | Official leaderboard | |
| 12 | | 77.3% | five-shot | Mar 2025 | Official leaderboard |
| 13 | | 77.3% | five-shot | Apr 2025 | Official leaderboard |
| 14 | | 76.3% | five-shot | Mar 2025 | Official leaderboard |
| 15 | Qwen2.5-32B-Instruct | 76% | five-shot | Official leaderboard | |
| 16 | | 75.7% | five-shot | Sep 2024 | Official leaderboard |
| 17 | EuroLLM-22B-Instruct-2512 | 75% | five-shot | Official leaderboard | |
| 18 | EuroLLM-22B | 74.1% | five-shot | Official leaderboard | |
| 19 | | 72.8% | five-shot | Apr 2025 | Official leaderboard |
| 20 | | 71.1% | five-shot | Sep 2024 | Official leaderboard |
| 21 | Llama-Krikri-8B-Base | 70.9% | five-shot | Official leaderboard | |
| 22 | Llama-Krikri-8B-Instruct | 70.8% | five-shot | Official leaderboard | |
| 23 | Plutus-8B-instruct | 70.8% | five-shot | Official leaderboard | |
| 24 | Qwen2.5-14B-Instruct | 70.6% | five-shot | Official leaderboard | |
| 25 | | 70.5% | zero-shot | Jan 2025 | Official leaderboard |
| 26 | GLM-4-9B | 70.2% | five-shot | Jun 2024 | Official leaderboard |
| 27 | GLM-4-9B-Chat | 68.8% | five-shot | Jun 2024 | Official leaderboard |
| 28 | EuroLLM-9B-Instruct | 68.5% | zero-shot | Official leaderboard | |
| 29 | qwen3-4b-instruct-2507 | 68.3% | five-shot | Aug 2025 | Official leaderboard |
| 30 | | 67.1% | five-shot | Apr 2025 | Official leaderboard |
| 31 | EuroLLM-9B | 66.3% | zero-shot | Official leaderboard | |
| 32 | | 65.6% | five-shot | Oct 2024 | Official leaderboard |
| 33 | Llama-3-8B-Instruct | 64.9% | five-shot | Official leaderboard | |
| 34 | | 64.8% | five-shot | Mar 2025 | Official leaderboard |
| 35 | Llama-3.1-8B-Instruct | 64.7% | five-shot | Official leaderboard | |
| 36 | | 64.5% | five-shot | Mar 2025 | Official leaderboard |
| 37 | | 64.2% | five-shot | Apr 2024 | Official leaderboard |
| 38 | Qwen2.5-7B-Instruct | 64.2% | five-shot | Official leaderboard | |
| 39 | Meltemi-7B-Instruct-v1.5 | 63.7% | five-shot | Official leaderboard | |
| 40 | | 63.3% | five-shot | Jul 2024 | Official leaderboard |
| 41 | Meltemi-7B-v1.5 | 63% | five-shot | Official leaderboard | |
| 42 | | 63% | five-shot | Sep 2024 | Official leaderboard |
| 43 | Aya-101 | 58.6% | five-shot | Official leaderboard | |
| 44 | Meltemi-7B-v1 | 58.4% | five-shot | Official leaderboard | |
| 45 | | 58% | five-shot | Apr 2025 | Official leaderboard |
| 46 | mT0-xxl | 56.6% | zero-shot | Official leaderboard | |
| 47 | Qwen2.5-3B-Instruct | 53.4% | five-shot | Official leaderboard | |
| 48 | Mistral-7B-Instruct-v0.3 | 51.6% | five-shot | Official leaderboard | |
| 49 | Llama-3.2-3B-Instruct | 50.5% | five-shot | Official leaderboard | |
| 50 | mT0-xl | 50% | zero-shot | Official leaderboard | |
| 51 | | 48.4% | five-shot | Sep 2024 | Official leaderboard |
| 52 | | 47.5% | five-shot | Sep 2024 | Official leaderboard |
| 53 | Qwen2.5-1.5B-Instruct | 46.5% | five-shot | Official leaderboard | |
| 54 | | 45.7% | five-shot | Mar 2025 | Official leaderboard |
| 55 | | 42.9% | five-shot | Sep 2024 | Official leaderboard |
| 56 | | 41% | five-shot | Apr 2025 | Official leaderboard |
| 57 | Llama-3.2-1B-Instruct | 40.5% | five-shot | Official leaderboard | |
| 58 | mT0-large | 37% | zero-shot | Official leaderboard | |
| 59 | Llama-2-7b-chat-hf | 36.8% | five-shot | Official leaderboard | |
| 60 | | 36.4% | zero-shot | Sep 2024 | Official leaderboard |
| 61 | | 35.3% | zero-shot | Jul 2023 | Official leaderboard |
| 62 | BLOOMZ-1b1 | 35.1% | zero-shot | Official leaderboard | |
| 63 | Qwen2.5-0.5B-Instruct | 34.9% | zero-shot | Official leaderboard | |
| 64 | | 34.7% | zero-shot | Sep 2024 | Official leaderboard |
| 65 | BLOOMZ-1b7 | 34.7% | zero-shot | Official leaderboard | |
| 66 | XGLM-4.5B | 33% | zero-shot | Official leaderboard | |
| 67 | XGLM-1.7B | 33% | zero-shot | Official leaderboard | |
| 68 | XGLM-2.9B | 32.7% | zero-shot | Official leaderboard | |
| 69 | BLOOMZ-7b1 | 32.4% | zero-shot | Official leaderboard | |
| 70 | EuroLLM-1.7B-Instruct | 32.4% | five-shot | Official leaderboard | |
| 71 | EuroLLM-1.7B | 32.3% | zero-shot | Official leaderboard | |
| 72 | XGLM-7.5B | 32.2% | zero-shot | Official leaderboard | |
| 73 | | 31.9% | zero-shot | Mar 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 93.2% | zero-shot | Dec 2025 | Official leaderboard |
| 2 | GPT-5.2 | 89.2% | five-shot | Dec 2025 | Official leaderboard |
| 3 | | 82.6% | five-shot | Mar 2025 | Official leaderboard |
| 4 | | 82.2% | five-shot | Sep 2024 | Official leaderboard |
| 5 | | 81.5% | five-shot | Dec 2024 | Official leaderboard |
| 6 | | 81.3% | five-shot | Mar 2025 | Official leaderboard |
| 7 | | 80.4% | five-shot | Jul 2024 | Official leaderboard |
| 8 | | 77.3% | five-shot | Mar 2025 | Official leaderboard |
| 9 | | 77.3% | five-shot | Apr 2025 | Official leaderboard |
| 10 | | 76.3% | five-shot | Mar 2025 | Official leaderboard |
| 11 | | 75.7% | five-shot | Sep 2024 | Official leaderboard |
| 12 | | 72.8% | five-shot | Apr 2025 | Official leaderboard |
| 13 | | 71.1% | five-shot | Sep 2024 | Official leaderboard |
| 14 | | 70.5% | zero-shot | Jan 2025 | Official leaderboard |
| 15 | GLM-4-9B | 70.2% | five-shot | Jun 2024 | Official leaderboard |
| 16 | GLM-4-9B-Chat | 68.8% | five-shot | Jun 2024 | Official leaderboard |
| 17 | qwen3-4b-instruct-2507 | 68.3% | five-shot | Aug 2025 | Official leaderboard |
| 18 | | 67.1% | five-shot | Apr 2025 | Official leaderboard |
| 19 | | 65.6% | five-shot | Oct 2024 | Official leaderboard |
| 20 | | 64.8% | five-shot | Mar 2025 | Official leaderboard |
| 21 | | 64.5% | five-shot | Mar 2025 | Official leaderboard |
| 22 | | 64.2% | five-shot | Apr 2024 | Official leaderboard |
| 23 | | 63.3% | five-shot | Jul 2024 | Official leaderboard |
| 24 | | 63% | five-shot | Sep 2024 | Official leaderboard |
| 25 | | 58% | five-shot | Apr 2025 | Official leaderboard |
| 26 | | 48.4% | five-shot | Sep 2024 | Official leaderboard |
| 27 | | 47.5% | five-shot | Sep 2024 | Official leaderboard |
| 28 | | 45.7% | five-shot | Mar 2025 | Official leaderboard |
| 29 | | 42.9% | five-shot | Sep 2024 | Official leaderboard |
| 30 | | 41% | five-shot | Apr 2025 | Official leaderboard |
| 31 | | 36.4% | zero-shot | Sep 2024 | Official leaderboard |
| 32 | | 35.3% | zero-shot | Jul 2023 | Official leaderboard |
| 33 | | 34.7% | zero-shot | Sep 2024 | Official leaderboard |
| 34 | | 31.9% | zero-shot | Mar 2025 | Official leaderboard |
Rankings by part
We read 5 parts from the benchmark’s paper, each ranking models on one part of the test, such as STEM, Humanities and Social Sciences. Gemini 3 Flash leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
STEM
34 models we track have a score here.
- 1
Gemini 3 Flash 92.8%
- 2 GPT-5.2 88.3%
- 3
Gemma 3 27B PT 82.3%
- 4
Qwen2.5-72B 82.2%
- 5
Gemma 3 27B ↑1 80.5%
- 6
Llama 3.3 70B Instruct ↓1 79.7%
- 7
Qwen3-14B ↑2 78.3%
- 8
Llama 3.1-70B ↓1 77.6%
- 9
Gemma 3 12B PT ↓1 77.2%
- 10
Qwen2.5-32B ↑1 75.6%
Humanities
34 models we track have a score here.
- 1
Gemini 3 Flash 92.9%
- 2 GPT-5.2 90.3%
- 3
Gemma 3 27B PT 82.9%
- 4
Qwen2.5-72B 82.8%
- 5
Llama 3.1-70B ↑2 82.8%
- 6
Llama 3.3 70B Instruct ↓1 82.5%
- 7
Gemma 3 27B ↓1 81.7%
- 8
Gemma 3 12B PT 78.6%
- 9
Gemma 3 12B ↑1 75.3%
- 10
Qwen2.5-32B ↑1 74.4%
Social Sciences
34 models we track have a score here.
- 1
Gemini 3 Flash 98.1%
- 2 GPT-5.2 90.7%
- 3
Llama 3.3 70B Instruct ↑2 84.2%
- 4
Gemma 3 27B PT ↓1 83.7%
- 5
Llama 3.1-70B ↑2 83.7%
- 6
Qwen2.5-72B ↓2 83.2%
- 7
Gemma 3 27B ↓1 83%
- 8
Gemma 3 12B ↑2 79.8%
- 9
Qwen3-14B 79.5%
- 10
Gemma 3 12B PT ↓2 78.3%
Other
34 models we track have a score here.
- 1
Gemini 3 Flash 91.8%
- 2 GPT-5.2 86.8%
- 3
Gemma 3 27B PT 80.7%
- 4
Llama 3.3 70B Instruct ↑1 79.1%
- 5
Qwen2.5-72B ↓1 78.6%
- 6
Gemma 3 27B 78.5%
- 7
Llama 3.1-70B 78.3%
- 8
Gemma 3 12B PT 76.2%
- 9
Gemma 3 12B ↑1 74.3%
- 10
Qwen3-14B ↓1 73.2%
Greek-specific
34 models we track have a score here.
- 1
Gemini 3 Flash 95.4%
- 2 GPT-5.2 93.8%
- 3
Llama 3.3 70B Instruct ↑2 87.9%
- 4
Llama 3.1-70B ↑3 87.3%
- 5
Gemma 3 27B PT ↓2 86.5%
- 6
Gemma 3 27B 86.1%
- 7
Qwen2.5-72B ↓3 85.9%
- 8
Gemma 3 12B ↑2 82.2%
- 9
Qwen3-14B 81.2%
- 10
Gemma 3 12B PT ↓2 80.7%
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 GreekMMLU paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| zero-shot | 93.2% | ||||
| five-shot | 91.2% | ||||
| GPT-5.2 | five-shot | 89.2% | |||
| GPT-5.2 | zero-shot | 87.8% | |||
| five-shot | 82.6% | ||||
| five-shot | 82.2% | ||||
| five-shot | 81.5% | ||||
| five-shot | 81.4% | ||||
| five-shot | 81.3% | ||||
| five-shot | 80.4% | ||||
| zero-shot | 79.7% | ||||
| zero-shot | 79.7% | ||||
| zero-shot | 79.4% | ||||
| zero-shot | 79.2% | ||||
| zero-shot | 78.9% | ||||
| five-shot | 77.3% | ||||
| five-shot | 77.3% | ||||
| five-shot | 76.3% | ||||
| five-shot | 75.7% | ||||
| zero-shot | 75.7% | ||||
| zero-shot | 75.3% | ||||
| zero-shot | 75% | ||||
| zero-shot | 73.1% | ||||
| five-shot | 72.8% | ||||
| five-shot | 71.1% | ||||
| zero-shot | 70.5% | ||||
| GLM-4-9B | five-shot | 70.2% | |||
| GLM-4-9B-Chat | five-shot | 68.8% | |||
| zero-shot | 68.8% | ||||
| qwen3-4b-instruct-2507 | five-shot | 68.3% | |||
| five-shot | 67.1% | ||||
| five-shot | 65.6% | ||||
| zero-shot | 65.3% | ||||
| GLM-4-9B | zero-shot | 65% | |||
| five-shot | 64.8% | ||||
| GLM-4-9B-Chat | zero-shot | 64.7% | |||
| five-shot | 64.5% | ||||
| qwen3-4b-instruct-2507 | zero-shot | 64.5% | |||
| zero-shot | 64.4% | ||||
| five-shot | 64.2% | ||||
| zero-shot | 64.2% | ||||
| zero-shot | 63.4% | ||||
| five-shot | 63.3% | ||||
| five-shot | 63% | ||||
| zero-shot | 62.2% | ||||
| five-shot | 58% | ||||
| zero-shot | 55.2% | ||||
| zero-shot | 54.1% | ||||
| zero-shot | 53.1% | ||||
| zero-shot | 52.2% | ||||
| zero-shot | 48.9% | ||||
| five-shot | 48.4% | ||||
| five-shot | 47.5% | ||||
| five-shot | 45.7% | ||||
| zero-shot | 45% | ||||
| zero-shot | 44.9% | ||||
| five-shot | 42.9% | ||||
| zero-shot | 42.8% | ||||
| five-shot | 41% | ||||
| zero-shot | 36.4% | ||||
| zero-shot | 35.9% | ||||
| zero-shot | 35.7% | ||||
| zero-shot | 35.3% | ||||
| five-shot | 35% | ||||
| zero-shot | 34.7% | ||||
| five-shot | 34.6% | ||||
| five-shot | 33.4% | ||||
| zero-shot | 31.9% | ||||
| five-shot | 30.8% |
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 GreekMMLU paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| zero-shot | 93.2% | ||||
| five-shot | 91.2% | ||||
| GPT-5.2 | five-shot | 89.2% | |||
| GPT-5.2 | zero-shot | 87.8% | |||
| five-shot | 82.6% | ||||
| five-shot | 82.2% | ||||
| five-shot | 81.5% | ||||
| five-shot | 81.4% | ||||
| five-shot | 81.3% | ||||
| five-shot | 80.4% | ||||
| zero-shot | 79.7% | ||||
| zero-shot | 79.7% | ||||
| zero-shot | 79.4% | ||||
| zero-shot | 79.2% | ||||
| zero-shot | 78.9% | ||||
| five-shot | 77.3% | ||||
| five-shot | 77.3% | ||||
| five-shot | 76.3% | ||||
| five-shot | 75.7% | ||||
| zero-shot | 75.7% | ||||
| zero-shot | 75.3% | ||||
| zero-shot | 75% | ||||
| zero-shot | 73.1% | ||||
| five-shot | 72.8% | ||||
| five-shot | 71.1% | ||||
| zero-shot | 70.5% | ||||
| GLM-4-9B | five-shot | 70.2% | |||
| GLM-4-9B-Chat | five-shot | 68.8% | |||
| zero-shot | 68.8% | ||||
| qwen3-4b-instruct-2507 | five-shot | 68.3% | |||
| five-shot | 67.1% | ||||
| five-shot | 65.6% | ||||
| zero-shot | 65.3% | ||||
| GLM-4-9B | zero-shot | 65% | |||
| five-shot | 64.8% | ||||
| GLM-4-9B-Chat | zero-shot | 64.7% | |||
| five-shot | 64.5% | ||||
| qwen3-4b-instruct-2507 | zero-shot | 64.5% | |||
| zero-shot | 64.4% | ||||
| five-shot | 64.2% | ||||
| zero-shot | 64.2% | ||||
| zero-shot | 63.4% | ||||
| five-shot | 63.3% | ||||
| five-shot | 63% | ||||
| zero-shot | 62.2% | ||||
| five-shot | 58% | ||||
| zero-shot | 55.2% | ||||
| zero-shot | 54.1% | ||||
| zero-shot | 53.1% | ||||
| zero-shot | 52.2% | ||||
| zero-shot | 48.9% | ||||
| five-shot | 48.4% | ||||
| five-shot | 47.5% | ||||
| five-shot | 45.7% | ||||
| zero-shot | 45% | ||||
| zero-shot | 44.9% | ||||
| five-shot | 42.9% | ||||
| zero-shot | 42.8% | ||||
| five-shot | 41% | ||||
| zero-shot | 36.4% | ||||
| zero-shot | 35.9% | ||||
| zero-shot | 35.7% | ||||
| zero-shot | 35.3% | ||||
| five-shot | 35% | ||||
| zero-shot | 34.7% | ||||
| five-shot | 34.6% | ||||
| five-shot | 33.4% | ||||
| zero-shot | 31.9% | ||||
| five-shot | 30.8% |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The GreekMMLU paper
GPT-5.2 on the official leaderboard passed 85% in December 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 GreekMMLU 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 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
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 benchmark’s paper (73 entries).
- Pass Models scored 34 models we track, plus 39 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 93.2%, past 85% since December 2025.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 69 rows for models we track and 39 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.