PerCoR
Models read the beginning of a Persian passage and choose its original ending from four options.
MIZAN reports accuracy for 45 model entries. The original test release contains 10,000 questions.
- Made by
- MCINext
- Released
- May 2025
- Kind
- Reasoning in many languages
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Language
- Persian
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- o3 #84 on the capability index
- Last model added
- September 2025
- Scores refreshed
- November 20th, 2026
Gemini 2.5 Pro leads with 94.3% among models released by September 2025 Gemini 2.5 Pro leads with 94.3% among models released by September 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. 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 September 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
- 94.3
Gemini 2.5 Pro
- 92.2 o3
- 91.2 Claude 3.7 Sonnet
- 88.4 GPT-4.1
- 87.2
Gemini 2.5 Flash
- 86.7 GPT-4o
- 86.5 GPT-5 mini
- 86.4
Gemini 2.0 Flash
- 82.5 deepseek-reasoner
- 82.4 deepseek-chat
- 81.4
Gemini 2.0 Flash-Lite
- 79.9
Cohere Command A
- 79.6
Llama 3.3 70B Instruct
- 78.1
Llama 4 Scout
- 77.1 gpt-oss-120b
- 77.1 GPT-4.1 mini
- 76.5
Qwen3-32B
- 76.3
Gemma 3 27B
- 76.0 GPT-4o mini
- 73.6
Command R+
- 73.1 GPT-5 nano
- 71.6 Claude 3.5 Haiku
- 70.9
Gemma 3 12B
- 69.7 gpt-oss-20b
- 69.6
Qwen3-14B
- 68.9
Mistral Small 3.1
- 68.8
Qwen3-30B-A3B
- 63.3
Aya Expanse 32B
- 60.0
Command R (03-2024)
- 59.1
Gemma 3n E4B
- 54.9 GPT-4.1 nano
- 54.4
Qwen3-8B
- 52.7 deepseek-r1-0528-qwen3-8b
- 50.3
Qwen3-4B
- 48.3
Gemma 3 4B
- 37.6 yandexgpt-lite
- 30.1 Mistral-7B-Instruct-v0.3
- 26.0
Gemma 3 1B
- 25.2 Llama-3.2-3B-Instruct
- 24.8
Gemma 3 270M
- 24.1 Llama-3.2-1B-Instruct
- 1
Gemini 2.5 Pro Jun 2025, Thinking 94.3%
- 2 o3 Thinking 92.2%
- 3 Claude 3.7 Sonnet Non-thinking 91.2%
- 4 GPT-4.1 Non-thinking 88.4%
- 5
Gemini 2.5 Flash Sep 2025, Thinking 87.2%
- 6 GPT-4o Mar 2025, Non-thinking 86.7%
- 7 GPT-5 mini Thinking 86.5%
- 8
Gemini 2.0 Flash Feb 2025, Non-thinking 86.4%
- 9 deepseek-reasoner Thinking 82.5%
- 10 deepseek-chat Non-thinking 82.4%
- 11
Gemini 2.0 Flash-Lite Non-thinking 81.4%
- 12
Cohere Command A Non-thinking 79.9%
- 13
Llama 3.3 70B Instruct Non-thinking 79.6%
- 14
Llama 4 Scout Non-thinking 78.1%
- 15 gpt-oss-120b Thinking 77.1%
- 16 GPT-4.1 mini Non-thinking 77.1%
- 17
Qwen3-32B Non-thinking 76.5%
- 18
Gemma 3 27B Non-thinking 76.3%
- 19 GPT-4o mini Non-thinking 76%
- 20
Command R+ Non-thinking 73.6%
- 21 GPT-5 nano Thinking 73.1%
- 22 Claude 3.5 Haiku Non-thinking 71.6%
- 23
Gemma 3 12B Non-thinking 70.9%
- 24 gpt-oss-20b Thinking 69.7%
- 25
Qwen3-14B Non-thinking 69.6%
- 26
Mistral Small 3.1 Non-thinking 68.9%
- 27
Qwen3-30B-A3B Non-thinking 68.8%
- 28
Aya Expanse 32B Non-thinking 63.3%
- 29
Command R (03-2024) Non-thinking 60%
- 30
Gemma 3n E4B Non-thinking 59.1%
- 31 GPT-4.1 nano Non-thinking 54.9%
- 32
Qwen3-8B Non-thinking 54.4%
- 33 deepseek-r1-0528-qwen3-8b Thinking 52.7%
- 34
Qwen3-4B Non-thinking 50.3%
- 35
Gemma 3 4B Non-thinking 48.3%
- 36 yandexgpt-lite Non-thinking 37.6%
- 37 Mistral-7B-Instruct-v0.3 Non-thinking 30.1%
- 38
Gemma 3 1B Non-thinking 26%
- 39 Llama-3.2-3B-Instruct Non-thinking 25.2%
- 40
Gemma 3 270M Non-thinking 24.8%
- 41 Llama-3.2-1B-Instruct Non-thinking 24.1%
Source: Official PerCoR leaderboard
- 94.3
Gemini 2.5 Pro
- 92.2 o3
- 91.2 Claude 3.7 Sonnet
- 88.4 GPT-4.1
- 87.2
Gemini 2.5 Flash
- 86.7 GPT-4o
- 86.5 GPT-5 mini
- 86.4
Gemini 2.0 Flash
- 81.4
Gemini 2.0 Flash-Lite
- 79.9
Cohere Command A
- 79.6
Llama 3.3 70B Instruct
- 78.1
Llama 4 Scout
- 77.1 gpt-oss-120b
- 77.1 GPT-4.1 mini
- 76.5
Qwen3-32B
- 76.3
Gemma 3 27B
- 76.0 GPT-4o mini
- 73.6
Command R+
- 73.1 GPT-5 nano
- 71.6 Claude 3.5 Haiku
- 70.9
Gemma 3 12B
- 69.7 gpt-oss-20b
- 69.6
Qwen3-14B
- 68.9
Mistral Small 3.1
- 68.8
Qwen3-30B-A3B
- 63.3
Aya Expanse 32B
- 60.0
Command R (03-2024)
- 59.1
Gemma 3n E4B
- 54.9 GPT-4.1 nano
- 54.4
Qwen3-8B
- 52.7 deepseek-r1-0528-qwen3-8b
- 50.3
Qwen3-4B
- 48.3
Gemma 3 4B
- 26.0
Gemma 3 1B
- 24.8
Gemma 3 270M
- 1
Gemini 2.5 Pro Jun 2025, Thinking 94.3%
- 2 o3 Thinking 92.2%
- 3 Claude 3.7 Sonnet Non-thinking 91.2%
- 4 GPT-4.1 Non-thinking 88.4%
- 5
Gemini 2.5 Flash Sep 2025, Thinking 87.2%
- 6 GPT-4o Mar 2025, Non-thinking 86.7%
- 7 GPT-5 mini Thinking 86.5%
- 8
Gemini 2.0 Flash Feb 2025, Non-thinking 86.4%
- 9
Gemini 2.0 Flash-Lite Non-thinking 81.4%
- 10
Cohere Command A Non-thinking 79.9%
- 11
Llama 3.3 70B Instruct Non-thinking 79.6%
- 12
Llama 4 Scout Non-thinking 78.1%
- 13 gpt-oss-120b Thinking 77.1%
- 14 GPT-4.1 mini Non-thinking 77.1%
- 15
Qwen3-32B Non-thinking 76.5%
- 16
Gemma 3 27B Non-thinking 76.3%
- 17 GPT-4o mini Non-thinking 76%
- 18
Command R+ Non-thinking 73.6%
- 19 GPT-5 nano Thinking 73.1%
- 20 Claude 3.5 Haiku Non-thinking 71.6%
- 21
Gemma 3 12B Non-thinking 70.9%
- 22 gpt-oss-20b Thinking 69.7%
- 23
Qwen3-14B Non-thinking 69.6%
- 24
Mistral Small 3.1 Non-thinking 68.9%
- 25
Qwen3-30B-A3B Non-thinking 68.8%
- 26
Aya Expanse 32B Non-thinking 63.3%
- 27
Command R (03-2024) Non-thinking 60%
- 28
Gemma 3n E4B Non-thinking 59.1%
- 29 GPT-4.1 nano Non-thinking 54.9%
- 30
Qwen3-8B Non-thinking 54.4%
- 31 deepseek-r1-0528-qwen3-8b Thinking 52.7%
- 32
Qwen3-4B Non-thinking 50.3%
- 33
Gemma 3 4B Non-thinking 48.3%
- 34
Gemma 3 1B Non-thinking 26%
- 35
Gemma 3 270M Non-thinking 24.8%
Source: Official PerCoR leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 94.3% | Thinking | Jun 2025 | Official leaderboard |
| 2 | o3 | 92.2% | Thinking | Apr 2025 | Official leaderboard |
| 3 | Claude 3.7 Sonnet | 91.2% | Non-thinking | Feb 2025 | Official leaderboard |
| 4 | GPT-4.1 | 88.4% | Non-thinking | Apr 2025 | Official leaderboard |
| 5 | | 87.2% | Thinking | Sep 2025 | Official leaderboard |
| 6 | GPT-4o Mar 2025 | 86.7% | Non-thinking | Mar 2025 | Official leaderboard |
| 7 | GPT-5 mini | 86.5% | Thinking | Aug 2025 | Official leaderboard |
| 8 | | 86.4% | Non-thinking | Feb 2025 | Official leaderboard |
| 9 | deepseek-reasoner open weights, 671B | 82.5% | Thinking | Official leaderboard | |
| 10 | deepseek-chat open weights, 671B | 82.4% | Non-thinking | Official leaderboard | |
| 11 | | 81.4% | Non-thinking | Feb 2025 | Official leaderboard |
| 12 | | 79.9% | Non-thinking | Mar 2025 | Official leaderboard |
| 13 | | 79.6% | Non-thinking | Dec 2024 | Official leaderboard |
| 14 | | 78.1% | Non-thinking | Apr 2025 | Official leaderboard |
| 15 | gpt-oss-120b | 77.1% | Thinking | Aug 2025 | Official leaderboard |
| 16 | GPT-4.1 mini | 77.1% | Non-thinking | Apr 2025 | Official leaderboard |
| 17 | | 76.5% | Non-thinking | Apr 2025 | Official leaderboard |
| 18 | | 76.3% | Non-thinking | Mar 2025 | Official leaderboard |
| 19 | GPT-4o mini | 76% | Non-thinking | Jul 2024 | Official leaderboard |
| 20 | | 73.6% | Non-thinking | Aug 2024 | Official leaderboard |
| 21 | GPT-5 nano | 73.1% | Thinking | Aug 2025 | Official leaderboard |
| 22 | Claude 3.5 Haiku | 71.6% | Non-thinking | Oct 2024 | Official leaderboard |
| 23 | | 70.9% | Non-thinking | Mar 2025 | Official leaderboard |
| 24 | gpt-oss-20b | 69.7% | Thinking | Aug 2025 | Official leaderboard |
| 25 | | 69.6% | Non-thinking | Apr 2025 | Official leaderboard |
| 26 | | 68.9% | Non-thinking | Mar 2025 | Official leaderboard |
| 27 | | 68.8% | Non-thinking | Apr 2025 | Official leaderboard |
| 28 | | 63.3% | Non-thinking | Oct 2024 | Official leaderboard |
| 29 | | 60% | Non-thinking | Mar 2024 | Official leaderboard |
| 30 | | 59.1% | Non-thinking | Jun 2025 | Official leaderboard |
| 31 | GPT-4.1 nano | 54.9% | Non-thinking | Apr 2025 | Official leaderboard |
| 32 | | 54.4% | Non-thinking | Apr 2025 | Official leaderboard |
| 33 | deepseek-r1-0528-qwen3-8b | 52.7% | Thinking | May 2025 | Official leaderboard |
| 34 | | 50.3% | Non-thinking | Apr 2025 | Official leaderboard |
| 35 | | 48.3% | Non-thinking | Mar 2025 | Official leaderboard |
| 36 | yandexgpt-lite open weights, 8B | 37.6% | Non-thinking | Official leaderboard | |
| 37 | Mistral-7B-Instruct-v0.3 open weights, 7.3B | 30.1% | Non-thinking | Official leaderboard | |
| 38 | | 26% | Non-thinking | Mar 2025 | Official leaderboard |
| 39 | Llama-3.2-3B-Instruct open weights, 3.2B | 25.2% | Non-thinking | Official leaderboard | |
| 40 | | 24.8% | Non-thinking | Aug 2025 | Official leaderboard |
| 41 | Llama-3.2-1B-Instruct open weights, 1.2B | 24.1% | Non-thinking | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 94.3% | Thinking | Jun 2025 | Official leaderboard |
| 2 | o3 | 92.2% | Thinking | Apr 2025 | Official leaderboard |
| 3 | Claude 3.7 Sonnet | 91.2% | Non-thinking | Feb 2025 | Official leaderboard |
| 4 | GPT-4.1 | 88.4% | Non-thinking | Apr 2025 | Official leaderboard |
| 5 | | 87.2% | Thinking | Sep 2025 | Official leaderboard |
| 6 | GPT-4o Mar 2025 | 86.7% | Non-thinking | Mar 2025 | Official leaderboard |
| 7 | GPT-5 mini | 86.5% | Thinking | Aug 2025 | Official leaderboard |
| 8 | | 86.4% | Non-thinking | Feb 2025 | Official leaderboard |
| 9 | | 81.4% | Non-thinking | Feb 2025 | Official leaderboard |
| 10 | | 79.9% | Non-thinking | Mar 2025 | Official leaderboard |
| 11 | | 79.6% | Non-thinking | Dec 2024 | Official leaderboard |
| 12 | | 78.1% | Non-thinking | Apr 2025 | Official leaderboard |
| 13 | gpt-oss-120b | 77.1% | Thinking | Aug 2025 | Official leaderboard |
| 14 | GPT-4.1 mini | 77.1% | Non-thinking | Apr 2025 | Official leaderboard |
| 15 | | 76.5% | Non-thinking | Apr 2025 | Official leaderboard |
| 16 | | 76.3% | Non-thinking | Mar 2025 | Official leaderboard |
| 17 | GPT-4o mini | 76% | Non-thinking | Jul 2024 | Official leaderboard |
| 18 | | 73.6% | Non-thinking | Aug 2024 | Official leaderboard |
| 19 | GPT-5 nano | 73.1% | Thinking | Aug 2025 | Official leaderboard |
| 20 | Claude 3.5 Haiku | 71.6% | Non-thinking | Oct 2024 | Official leaderboard |
| 21 | | 70.9% | Non-thinking | Mar 2025 | Official leaderboard |
| 22 | gpt-oss-20b | 69.7% | Thinking | Aug 2025 | Official leaderboard |
| 23 | | 69.6% | Non-thinking | Apr 2025 | Official leaderboard |
| 24 | | 68.9% | Non-thinking | Mar 2025 | Official leaderboard |
| 25 | | 68.8% | Non-thinking | Apr 2025 | Official leaderboard |
| 26 | | 63.3% | Non-thinking | Oct 2024 | Official leaderboard |
| 27 | | 60% | Non-thinking | Mar 2024 | Official leaderboard |
| 28 | | 59.1% | Non-thinking | Jun 2025 | Official leaderboard |
| 29 | GPT-4.1 nano | 54.9% | Non-thinking | Apr 2025 | Official leaderboard |
| 30 | | 54.4% | Non-thinking | Apr 2025 | Official leaderboard |
| 31 | deepseek-r1-0528-qwen3-8b | 52.7% | Thinking | May 2025 | Official leaderboard |
| 32 | | 50.3% | Non-thinking | Apr 2025 | Official leaderboard |
| 33 | | 48.3% | Non-thinking | Mar 2025 | Official leaderboard |
| 34 | | 26% | Non-thinking | Mar 2025 | Official leaderboard |
| 35 | | 24.8% | Non-thinking | Aug 2025 | Official leaderboard |
Rankings by part
We read 44 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as acc strict, donyaeeqtesad acc and isna acc. Gemini 2.5 Pro leads 34 of them, and 5 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
acc strict
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 94.3%
- 2 o3 92.2%
- 3 Claude 3.7 Sonnet 91.2%
- 4 GPT-4.1 88.4%
- 5
Gemini 2.5 Flash Sep 2025 87.1%
- 6 GPT-5 mini ↑1 86.5%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 86.4%
- 8
Gemini 2.0 Flash-Lite ↑1 81.4%
- 9
Cohere Command A ↑1 79.8%
- 10 GPT-4o Mar 2025 ↓4 78.3%
donyaeeqtesad acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 94.5%
- 2 o3 92.1%
- 3 Claude 3.7 Sonnet 91.8%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 87%
- 5 GPT-4.1 ↓1 86.1%
- 6 GPT-4o Mar 2025 84.6%
- 7 GPT-5 mini 84.3%
- 8
Gemini 2.0 Flash Feb 2025 84%
- 9
Gemini 2.0 Flash-Lite 79.8%
- 10
Llama 4 Scout ↑2 79.5%
isna acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 94.9%
- 2 Claude 3.7 Sonnet ↑1 90.6%
- 3 o3 ↓1 90.2%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 86.7%
- 5 GPT-4o Mar 2025 ↑1 83.6%
- 6 GPT-4.1 ↓2 83.6%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 82.8%
- 8 GPT-5 mini ↓1 80.9%
- 9
Gemini 2.0 Flash-Lite 77.7%
- 10
Cohere Command A 77%
ninisite article acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 93.7%
- 2 Claude 3.7 Sonnet ↑1 90.3%
- 3 o3 ↓1 90.3%
- 4 GPT-5 mini ↑3 87.6%
- 5 GPT-4.1 ↓1 87%
- 6
Gemini 2.5 Flash Sep 2025 ↓1 86.2%
- 7 GPT-4o Mar 2025 ↓1 85.9%
- 8
Gemini 2.0 Flash Feb 2025 82.4%
- 9
Gemini 2.0 Flash-Lite 80%
- 10 gpt-oss-120b ↑3 77.8%
virgool 4 acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 96%
- 2 o3 93.4%
- 3 Claude 3.7 Sonnet 92.4%
- 4 GPT-4.1 90.3%
- 5
Gemini 2.5 Flash Sep 2025 90.3%
- 6 GPT-5 mini ↑1 89.1%
- 7 GPT-4o Mar 2025 ↓1 88.8%
- 8
Gemini 2.0 Flash Feb 2025 88.5%
- 9
Llama 3.3 70B Instruct ↑2 85.5%
- 10
Cohere Command A 84%
khabaronline acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 93.9%
- 2 Claude 3.7 Sonnet ↑1 93.6%
- 3 o3 ↓1 92%
- 4 GPT-4o Mar 2025 ↑2 89.6%
- 5 GPT-4.1 ↓1 88.4%
- 6 GPT-5 mini ↑1 86.4%
- 7
Gemini 2.5 Flash Sep 2025 ↓2 85.2%
- 8
Gemini 2.0 Flash Feb 2025 84%
- 9
Gemini 2.0 Flash-Lite 79.6%
- 10
Llama 3.3 70B Instruct ↑1 78%
digiato acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 94.9%
- 2 o3 91.9%
- 3 Claude 3.7 Sonnet 91.8%
- 4 GPT-5 mini ↑3 89.6%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 87.5%
- 6
Gemini 2.5 Flash Sep 2025 ↓1 87.5%
- 7 GPT-4.1 ↓3 87.1%
- 8 GPT-4o Mar 2025 ↓2 85%
- 9
Gemini 2.0 Flash-Lite 80.4%
- 10
Qwen3-32B ↑5 80.4%
doctoreto acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 92.4%
- 2 o3 92%
- 3 GPT-5 mini ↑4 91%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 90%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 89.5%
- 6 Claude 3.7 Sonnet ↓3 89%
- 7 GPT-4.1 ↓3 88.5%
- 8 GPT-4o Mar 2025 ↓2 85.5%
- 9
Gemini 2.0 Flash-Lite 82%
- 10
Cohere Command A 82%
sarzamindownload acc
35 models we track have a score here.
- 1 o3 ↑1 93.5%
- 2
Gemini 2.5 Pro Jun 2025 ↓1 89.3%
- 3 Claude 3.7 Sonnet 87.6%
- 4 GPT-4.1 85.6%
- 5
Gemini 2.5 Flash Sep 2025 85.6%
- 6
Gemini 2.0 Flash Feb 2025 ↑2 83.7%
- 7 GPT-5 mini 83.7%
- 8 GPT-4o Mar 2025 ↓2 82.4%
- 9 gpt-oss-120b ↑4 82.4%
- 10 gpt-oss-20b ↑12 79.7%
hamgardi acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 91.4%
- 2 o3 88.8%
- 3 Claude 3.7 Sonnet 87%
- 4 GPT-4.1 83.8%
- 5
Gemini 2.5 Flash Sep 2025 82.3%
- 6 GPT-4o Mar 2025 82%
- 7 GPT-5 mini 82%
- 8
Gemini 2.0 Flash Feb 2025 80.2%
- 9
Gemini 2.0 Flash-Lite 76.7%
- 10
Cohere Command A 76.1%
bigbangpage acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 98%
- 2 o3 94.9%
- 3 Claude 3.7 Sonnet 93.6%
- 4 GPT-4.1 92.4%
- 5 GPT-4o Mar 2025 ↑1 91.7%
- 6 GPT-5 mini ↑1 90.4%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 89.8%
- 8
Gemini 2.5 Flash Sep 2025 ↓3 89.8%
- 9
Cohere Command A ↑1 87.9%
- 10
Gemini 2.0 Flash-Lite ↓1 86%
wiki ahlolbait acc
35 models we track have a score here.
- 1 Claude 3.7 Sonnet ↑2 96.7%
- 2
Gemini 2.5 Pro Jun 2025 ↓1 96%
- 3 o3 ↓1 95.4%
- 4
Gemini 2.0 Flash Feb 2025 ↑4 94.1%
- 5 GPT-4o Mar 2025 ↑1 93.4%
- 6 GPT-4.1 ↓2 92.8%
- 7
Gemini 2.0 Flash-Lite ↑2 89.5%
- 8 GPT-5 mini ↓1 89.5%
- 9
Gemini 2.5 Flash Sep 2025 ↓4 88.8%
- 10 Claude 3.5 Haiku ↑10 86.8%
virgool 3 acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 96.1%
- 2 o3 93.7%
- 3 Claude 3.7 Sonnet 92.2%
- 4 GPT-4.1 91%
- 5
Gemini 2.5 Flash Sep 2025 91%
- 6
Gemini 2.0 Flash Feb 2025 ↑2 89.6%
- 7 GPT-4o Mar 2025 ↓1 86.6%
- 8 GPT-5 mini ↓1 86.3%
- 9
Gemini 2.0 Flash-Lite 83%
- 10
Cohere Command A 81.2%
virgool 2 acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 95.7%
- 2 o3 94.2%
- 3 Claude 3.7 Sonnet 92.4%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 89.3%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 89%
- 6 GPT-5 mini ↑1 88.7%
- 7 GPT-4o Mar 2025 ↓1 88.4%
- 8 GPT-4.1 ↓4 88.4%
- 9
Gemini 2.0 Flash-Lite 84.1%
- 10
Llama 3.3 70B Instruct ↑1 82.9%
virgool 1 acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 95.2%
- 2 o3 92.4%
- 3 Claude 3.7 Sonnet 91.5%
- 4 GPT-5 mini ↑3 89.2%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 88.9%
- 6 GPT-4o Mar 2025 88.9%
- 7
Gemini 2.5 Flash Sep 2025 ↓2 87.7%
- 8 GPT-4.1 ↓4 86.1%
- 9
Llama 3.3 70B Instruct ↑2 82.6%
- 10
Gemini 2.0 Flash-Lite ↓1 82%
hamshahrionline acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 95.7%
- 2 Claude 3.7 Sonnet ↑1 94.4%
- 3 GPT-4o Mar 2025 ↑3 94.1%
- 4 o3 ↓2 92.8%
- 5 GPT-4.1 ↓1 91.1%
- 6 GPT-5 mini ↑1 87.9%
- 7
Gemini 2.5 Flash Sep 2025 ↓2 87.9%
- 8
Gemini 2.0 Flash Feb 2025 87.5%
- 9
Llama 3.3 70B Instruct ↑2 86.2%
- 10
Gemini 2.0 Flash-Lite ↓1 85.6%
tabnak acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 95.4%
- 2 o3 92.9%
- 3 Claude 3.7 Sonnet 92.5%
- 4 GPT-4o Mar 2025 ↑2 88.8%
- 5 GPT-4.1 ↓1 88.8%
- 6 GPT-5 mini ↑1 88.4%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 87.6%
- 8
Gemini 2.5 Flash Sep 2025 ↓3 87.6%
- 9
Cohere Command A ↑1 83%
- 10
Llama 3.3 70B Instruct ↑1 82.6%
alibaba acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 93.7%
- 2
Gemini 2.5 Flash Sep 2025 ↑3 91.8%
- 3 Claude 3.7 Sonnet 91.2%
- 4 GPT-4.1 91.2%
- 5 o3 ↓3 91.2%
- 6
Cohere Command A ↑4 85.6%
- 7 GPT-4o Mar 2025 ↓1 85.3%
- 8 GPT-5 mini ↓1 84.6%
- 9
Gemini 2.0 Flash Feb 2025 ↓1 84.3%
- 10
Llama 3.3 70B Instruct ↑1 83.7%
digikala mag acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 96.9%
- 2 o3 94.6%
- 3 Claude 3.7 Sonnet 92%
- 4 GPT-4.1 91.2%
- 5
Gemini 2.5 Flash Sep 2025 89.8%
- 6 GPT-5 mini ↑1 89.2%
- 7 GPT-4o Mar 2025 ↓1 88.2%
- 8
Gemini 2.0 Flash Feb 2025 87.8%
- 9 gpt-oss-120b ↑4 82.4%
- 10
Gemini 2.0 Flash-Lite ↓1 81.8%
yjc acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 93%
- 2 Claude 3.7 Sonnet ↑1 90.8%
- 3 o3 ↓1 90.8%
- 4 GPT-4.1 86.8%
- 5
Gemini 2.5 Flash Sep 2025 85.6%
- 6 GPT-5 mini ↑1 82.2%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 81.6%
- 8 GPT-4o Mar 2025 ↓2 81%
- 9
Gemini 2.0 Flash-Lite 80.5%
- 10 GPT-4o mini ↑7 76.4%
beytoote acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 95.6%
- 2 o3 93.4%
- 3 GPT-4.1 ↑1 92%
- 4 Claude 3.7 Sonnet ↓1 91.7%
- 5 GPT-4o Mar 2025 ↑1 88.9%
- 6
Gemini 2.5 Flash Sep 2025 ↓1 88.6%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 88.4%
- 8 GPT-5 mini ↓1 88.4%
- 9
Gemini 2.0 Flash-Lite 81.2%
- 10 gpt-oss-120b ↑3 80.3%
asriran acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 92.2%
- 2 o3 91.3%
- 3
Gemini 2.0 Flash Feb 2025 ↑5 88.9%
- 4 Claude 3.7 Sonnet ↓1 88.9%
- 5 GPT-4.1 ↓1 87.9%
- 6 GPT-4o Mar 2025 85.5%
- 7 GPT-5 mini 83.6%
- 8
Gemini 2.5 Flash Sep 2025 ↓3 82.1%
- 9
Cohere Command A ↑1 79.2%
- 10
Llama 3.3 70B Instruct ↑1 78.7%
ecoiran acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 91.6%
- 2 Claude 3.7 Sonnet ↑1 88.3%
- 3 o3 ↓1 88.3%
- 4 GPT-4.1 86.3%
- 5
Gemini 2.5 Flash Sep 2025 84.8%
- 6 GPT-5 mini ↑1 83.8%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 81.3%
- 8 GPT-4o Mar 2025 ↓2 81.3%
- 9
Gemini 2.0 Flash-Lite 76.8%
- 10
Llama 4 Scout ↑2 75.2%
hawzah acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 97.5%
- 2 o3 95.8%
- 3 Claude 3.7 Sonnet 93.8%
- 4 GPT-4.1 92.7%
- 5 GPT-4o Mar 2025 ↑1 91.3%
- 6
Gemini 2.5 Flash Sep 2025 ↓1 89.6%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 89%
- 8 GPT-5 mini ↓1 86.2%
- 9
Gemini 2.0 Flash-Lite 85.1%
- 10
Llama 3.3 70B Instruct ↑1 84.6%
zoomit acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 95.9%
- 2 o3 95.8%
- 3 Claude 3.7 Sonnet 93.4%
- 4 GPT-4.1 91.5%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 90.8%
- 6 GPT-4o Mar 2025 90.8%
- 7
Gemini 2.5 Flash Sep 2025 ↓2 90.8%
- 8 GPT-5 mini ↓1 89.7%
- 9
Gemini 2.0 Flash-Lite 84.7%
- 10
Cohere Command A 84.3%
wikipedia acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 98.5%
- 2 o3 97.6%
- 3 Claude 3.7 Sonnet 94.8%
- 4 GPT-4.1 94.3%
- 5 GPT-4o Mar 2025 ↑1 92.9%
- 6
Gemini 2.5 Flash Sep 2025 ↓1 92.4%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 91.9%
- 8 GPT-5 mini ↓1 91.4%
- 9
Cohere Command A ↑1 91%
- 10
Llama 3.3 70B Instruct ↑1 90.5%
namnak acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 91.7%
- 2 o3 90.2%
- 3 Claude 3.7 Sonnet 88.8%
- 4 GPT-5 mini ↑3 86.1%
- 5 GPT-4.1 ↓1 85.8%
- 6 GPT-4o Mar 2025 84.5%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 83.9%
- 8
Gemini 2.5 Flash Sep 2025 ↓3 83.1%
- 9
Llama 3.3 70B Instruct ↑2 78.7%
- 10
Gemini 2.0 Flash-Lite ↓1 78.5%
khodro45 acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 92.5%
- 2 o3 91.9%
- 3 Claude 3.7 Sonnet 91.2%
- 4
Gemini 2.0 Flash Feb 2025 ↑4 88.2%
- 5 GPT-5 mini ↑2 87.5%
- 6 GPT-4.1 ↓2 86%
- 7 GPT-4o Mar 2025 ↓1 84.6%
- 8
Gemini 2.0 Flash-Lite ↑1 83.8%
- 9 gpt-oss-120b ↑4 83.8%
- 10
Gemini 2.5 Flash Sep 2025 ↓5 83.8%
fidibo acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 95.1%
- 2 o3 94.7%
- 3
Gemini 2.5 Flash Sep 2025 ↑2 92.1%
- 4 Claude 3.7 Sonnet ↓1 91.6%
- 5 GPT-4.1 ↓1 90.7%
- 6
Gemini 2.0 Flash Feb 2025 ↑2 90.3%
- 7 GPT-4o Mar 2025 ↓1 87.2%
- 8 GPT-5 mini ↓1 85.5%
- 9
Gemini 2.0 Flash-Lite 84.6%
- 10
Cohere Command A 84.6%
newmiind acc
35 models we track have a score here.
- 1 o3 ↑1 87.8%
- 2
Gemini 2.5 Pro Jun 2025 ↓1 87.8%
- 3 Claude 3.7 Sonnet 82.6%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 81.9%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 81.3%
- 6 GPT-4o Mar 2025 81.3%
- 7 GPT-4.1 ↓3 80.2%
- 8 GPT-5 mini ↓1 80.2%
- 9
Gemini 2.0 Flash-Lite 77.1%
- 10
Llama 3.3 70B Instruct ↑1 74.7%
taaghche acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 96.8%
- 2 o3 94.9%
- 3
Gemini 2.0 Flash Feb 2025 ↑5 94.2%
- 4 Claude 3.7 Sonnet ↓1 93.6%
- 5 GPT-5 mini ↑2 91.7%
- 6 GPT-4o Mar 2025 90.4%
- 7
Gemini 2.5 Flash Sep 2025 ↓2 90.4%
- 8 GPT-4.1 ↓4 89.7%
- 9
Gemini 2.0 Flash-Lite 85.3%
- 10
Cohere Command A 84%
motamem acc
35 models we track have a score here.
- 1 o3 ↑1 98.9%
- 2 Claude 3.7 Sonnet ↑1 96.8%
- 3
Gemini 2.5 Pro Jun 2025 ↓2 96.8%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 93.7%
- 5 GPT-4.1 ↓1 92.6%
- 6
Gemini 2.0 Flash Feb 2025 ↑2 91.6%
- 7
Gemini 2.0 Flash-Lite ↑2 91.6%
- 8 GPT-4o Mar 2025 ↓2 91.6%
- 9 GPT-5 mini ↓2 90.5%
- 10
Llama 3.3 70B Instruct ↑1 89.5%
varzesh3 acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 96.5%
- 2 o3 95.7%
- 3 Claude 3.7 Sonnet 94%
- 4 GPT-4.1 91%
- 5 GPT-4o Mar 2025 ↑1 90%
- 6 GPT-5 mini ↑1 89.6%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 89.3%
- 8
Gemini 2.5 Flash Sep 2025 ↓3 88.3%
- 9
Gemini 2.0 Flash-Lite 83.9%
- 10
Llama 4 Scout ↑2 81.6%
mehrnews acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 93.1%
- 2 Claude 3.7 Sonnet ↑1 92.3%
- 3 o3 ↓1 89.9%
- 4 GPT-4.1 85.1%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 84.3%
- 6 GPT-5 mini ↑1 83.9%
- 7 GPT-4o Mar 2025 ↓1 81.5%
- 8
Gemini 2.5 Flash Sep 2025 ↓3 80.6%
- 9
Gemini 2.0 Flash-Lite 74.6%
- 10
Llama 3.3 70B Instruct ↑1 74.2%
tasnim acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 93.7%
- 2 Claude 3.7 Sonnet ↑1 91.2%
- 3 o3 ↓1 88.8%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 86.9%
- 5 GPT-4o Mar 2025 ↑1 84.6%
- 6 GPT-5 mini ↑1 84.2%
- 7 GPT-4.1 ↓3 83.1%
- 8
Gemini 2.0 Flash-Lite ↑1 82.3%
- 9
Gemini 2.0 Flash Feb 2025 ↓1 81.5%
- 10
Llama 4 Scout ↑2 77.7%
magerta acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 86.8%
- 2 Claude 3.7 Sonnet ↑1 85.7%
- 3 GPT-4.1 ↑1 84%
- 4 o3 ↓2 84%
- 5 GPT-4o Mar 2025 ↑1 78.2%
- 6 GPT-5 mini ↑1 78.2%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 77.7%
- 8 gpt-oss-120b ↑5 76.9%
- 9
Gemini 2.5 Flash Sep 2025 ↓4 76%
- 10
Gemini 2.0 Flash-Lite ↓1 74.8%
radiokodak book acc
35 models we track have a score here.
- 1 Claude 3.7 Sonnet ↑2 87%
- 2 GPT-4.1 ↑2 87%
- 3 o3 ↓1 82.6%
- 4
Gemini 2.5 Flash Sep 2025 ↑1 82.6%
- 5
Gemini 2.5 Pro Jun 2025 ↓4 81.8%
- 6 GPT-4o Mar 2025 78.3%
- 7 GPT-5 mini 73.9%
- 8
Gemini 2.0 Flash Feb 2025 69.6%
- 9
Gemini 2.0 Flash-Lite 69.6%
- 10
Gemma 3 27B ↑6 69.6%
vipofilm acc
35 models we track have a score here.
- 1
Gemini 2.0 Flash Feb 2025 ↑7 100%
- 2
Gemini 2.0 Flash-Lite ↑7 100%
- 3 Claude 3.7 Sonnet 100%
- 4 GPT-4.1 100%
- 5
Gemini 2.5 Pro Jun 2025 ↓4 100%
- 6 gpt-oss-120b ↑7 100%
- 7 GPT-5 mini 100%
- 8
Command R+ ↑10 92.3%
- 9 Claude 3.5 Haiku ↑11 92.3%
- 10
Cohere Command A 92.3%
wikishia acc
35 models we track have a score here.
- 1
Gemini 2.0 Flash-Lite ↑8 100%
- 2 Claude 3.7 Sonnet ↑1 100%
- 3 GPT-4.1 ↑1 100%
- 4
Gemini 2.5 Pro Jun 2025 ↓3 100%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 97%
- 6 GPT-4o Mar 2025 97%
- 7 o3 ↓5 97%
- 8
Gemma 3 27B ↑8 93.9%
- 9
Gemini 2.5 Flash Sep 2025 ↓4 93.9%
- 10 GPT-5 mini ↓3 90.9%
voolak acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 95.2%
- 2 GPT-4.1 ↑2 88.4%
- 3 Claude 3.7 Sonnet 86%
- 4 o3 ↓2 83.7%
- 5
Gemini 2.5 Flash Sep 2025 83.7%
- 6
Llama 3.3 70B Instruct ↑5 79.1%
- 7
Gemini 2.0 Flash-Lite ↑2 79.1%
- 8 GPT-5 mini ↓1 79.1%
- 9 GPT-4o Mar 2025 ↓3 76.7%
- 10 GPT-4o mini ↑7 74.4%
farsroid acc
35 models we track have a score here.
- 1 GPT-4.1 ↑3 84.2%
- 2 GPT-5 mini ↑5 84.2%
- 3
Gemini 2.5 Pro Jun 2025 ↓2 83.3%
- 4 o3 ↓2 81.6%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 78.9%
- 6 Claude 3.7 Sonnet ↓3 78.9%
- 7
Gemini 2.5 Flash Sep 2025 ↓2 78.9%
- 8 GPT-4o mini ↑9 76.3%
- 9
Llama 4 Scout ↑3 76.3%
- 10
Gemma 3 12B ↑11 73.7%
parsiday acc
35 models we track have a score here.
- 1
Gemini 2.5 Pro Jun 2025 93.2%
- 2 o3 89.2%
- 3 GPT-4.1 ↑1 88.3%
- 4 GPT-4o Mar 2025 ↑2 85%
- 5 Claude 3.7 Sonnet ↓2 83.3%
- 6 GPT-5 mini ↑1 82.5%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 81.7%
- 8
Gemini 2.5 Flash Sep 2025 ↓3 77.5%
- 9
Llama 3.3 70B Instruct ↑2 75.8%
- 10
Gemini 2.0 Flash-Lite ↓1 74.2%
soft98 acc
35 models we track have a score here.
- 1
Gemini 2.0 Flash Feb 2025 ↑7 100%
- 2 Claude 3.7 Sonnet ↑1 100%
- 3 o3 ↓1 100%
- 4
Command R+ ↑14 90%
- 5 Claude 3.5 Haiku ↑15 90%
- 6
Llama 3.3 70B Instruct ↑5 90%
- 7
Gemini 2.0 Flash-Lite ↑2 90%
- 8
Gemma 3 4B ↑25 90%
- 9 GPT-4.1 ↓5 90%
- 10
Qwen3-14B ↑13 90%
ninisite discussion acc
35 models we track have a score here.
- 1
Gemini 2.0 Flash Feb 2025 ↑7 90%
- 2 GPT-4o Mar 2025 ↑4 90%
- 3 GPT-4.1 ↑1 90%
- 4
Gemini 2.5 Pro Jun 2025 ↓3 90%
- 5 Claude 3.7 Sonnet ↓2 80%
- 6
Gemma 3 12B ↑15 80%
- 7
Gemini 2.5 Flash Sep 2025 ↓2 80%
- 8
Command R+ ↑10 70%
- 9
Gemma 3 27B ↑7 70%
- 10
Qwen3-14B ↑13 70%
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: Official PerCoR leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Thinking | 94.3% | ||||
| o3 | Thinking | 92.2% | |||
| Claude 3.7 Sonnet | Non-thinking | 91.2% | |||
| GPT-4.1 | Non-thinking | 88.4% | |||
| Thinking | 87.2% | ||||
| GPT-4o Mar 2025 | Non-thinking | 86.7% | |||
| GPT-5 mini | Thinking | 86.5% | |||
| Non-thinking | 86.4% | ||||
| Non-thinking | 81.4% | ||||
| Non-thinking | 79.9% | ||||
| Non-thinking | 79.6% | ||||
| Non-thinking | 78.1% | ||||
| gpt-oss-120b | Thinking | 77.1% | |||
| GPT-4.1 mini | Non-thinking | 77.1% | |||
| GPT-5 mini | minimal reasoning | 76.9% | |||
| Non-thinking | 76.5% | ||||
| Non-thinking | 76.3% | ||||
| GPT-4o mini | Non-thinking | 76% | |||
| Non-thinking | 73.6% | ||||
| GPT-5 nano | Thinking | 73.1% | |||
| Claude 3.5 Haiku | Non-thinking | 71.6% | |||
| gpt-oss-120b | low reasoning | 71.2% | |||
| Non-thinking | 70.9% | ||||
| gpt-oss-20b | Thinking | 69.7% | |||
| Non-thinking | 69.6% | ||||
| Non-thinking | 68.9% | ||||
| Non-thinking | 68.8% | ||||
| gpt-oss-20b | low reasoning | 64.5% | |||
| Non-thinking | 63.3% | ||||
| Non-thinking | 60% | ||||
| Non-thinking | 59.1% | ||||
| GPT-4.1 nano | Non-thinking | 54.9% | |||
| Non-thinking | 54.4% | ||||
| deepseek-r1-0528-qwen3-8b | Thinking | 52.7% | |||
| GPT-5 nano | minimal reasoning | 51.9% | |||
| Non-thinking | 50.3% | ||||
| Non-thinking | 48.3% | ||||
| Non-thinking | 26% | ||||
| Non-thinking | 24.8% |
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: Official PerCoR leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Thinking | 94.3% | ||||
| o3 | Thinking | 92.2% | |||
| Claude 3.7 Sonnet | Non-thinking | 91.2% | |||
| GPT-4.1 | Non-thinking | 88.4% | |||
| Thinking | 87.2% | ||||
| GPT-4o Mar 2025 | Non-thinking | 86.7% | |||
| GPT-5 mini | Thinking | 86.5% | |||
| Non-thinking | 86.4% | ||||
| Non-thinking | 81.4% | ||||
| Non-thinking | 79.9% | ||||
| Non-thinking | 79.6% | ||||
| Non-thinking | 78.1% | ||||
| gpt-oss-120b | Thinking | 77.1% | |||
| GPT-4.1 mini | Non-thinking | 77.1% | |||
| GPT-5 mini | minimal reasoning | 76.9% | |||
| Non-thinking | 76.5% | ||||
| Non-thinking | 76.3% | ||||
| GPT-4o mini | Non-thinking | 76% | |||
| Non-thinking | 73.6% | ||||
| GPT-5 nano | Thinking | 73.1% | |||
| Claude 3.5 Haiku | Non-thinking | 71.6% | |||
| gpt-oss-120b | low reasoning | 71.2% | |||
| Non-thinking | 70.9% | ||||
| gpt-oss-20b | Thinking | 69.7% | |||
| Non-thinking | 69.6% | ||||
| Non-thinking | 68.9% | ||||
| Non-thinking | 68.8% | ||||
| gpt-oss-20b | low reasoning | 64.5% | |||
| Non-thinking | 63.3% | ||||
| Non-thinking | 60% | ||||
| Non-thinking | 59.1% | ||||
| GPT-4.1 nano | Non-thinking | 54.9% | |||
| Non-thinking | 54.4% | ||||
| deepseek-r1-0528-qwen3-8b | Thinking | 52.7% | |||
| GPT-5 nano | minimal reasoning | 51.9% | |||
| Non-thinking | 50.3% | ||||
| Non-thinking | 48.3% | ||||
| Non-thinking | 26% | ||||
| Non-thinking | 24.8% |
2 models have held number one on the official leaderboard since its release in May 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official PerCoR leaderboard
Gemini 2.0 Flash on the official leaderboard passed 85% in February 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 PerCoR 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 official leaderboard (41 entries).
- Pass Models scored 35 models we track, plus 6 other entries on its leaderboard.
- Note Newest score The newest came 14 months ago, in September 2025, for Gemini 2.5 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 o3 (#84).
- Note Headroom Saturated. The best is 94.3%, past 85% since February 2025.
Leaderboard import 3 pass
- Pass Importer We read the board on mcinext-mizan-llm-leaderboard.hf.space: 39 rows for models we track and 6 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.