Nejumi LLM Leaderboard 3
Version 3
Models complete Japanese language and alignment evaluations. The organizers group results by use case and include safety evaluations.
The project announced a new scoring structure with this release.
- Made by
- Weights & Biases Japan
- Released
- July 2024
- Kind
- Reading in many languages
- Level
- School
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Subject
- Languages and literature
- Language
- Japanese
- Interaction
- Single turn
- Refresh
- New versions replace it
- Run by
- Anyone
- Based on
- Nejumi LLM Leaderboard 2
- Strongest model tested
- o3 #84 on the capability index
- Last model added
- August 2025
- Scores refreshed
- November 20th, 2026
o3 leads with 0.83 among models released by August 2025 o3 leads with 0.83 among models released by August 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 August 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
- 0.83 o3
- 0.83 Claude Sonnet 4
- 0.82 o1
- 0.82 Claude 3.7 Sonnet
- 0.82 Claude 3.5 Sonnet
- 0.81 Claude Opus 4
- 0.81 Claude 3.5 Sonnet (October 2024)
- 0.80 GPT-4.5
- 0.80 o1-preview
- 0.80 o4-mini
- 0.79 chatgpt-4o-latest
- 0.79 GPT-4o
- 0.79 Grok 3
- 0.79
Gemini 2.5 Flash
- 0.79
Qwen2.5-Max
- 0.79 GPT-4.1
- 0.78 GPT-4o
- 0.78
Gemini 1.5 Pro
- 0.78 o3-mini
- 0.78 deepseek-chat
- 0.78
Gemini 2.0 Flash
- 0.77 GPT-4o
- 0.77
Gemini 2.5 Pro
- 0.77 gpt-oss-120b
- 0.77
DeepSeek-R1
- 0.77
Cohere Command A
- 0.76
Athene-V2-Chat
- 0.76 GPT-4 Turbo
- 0.76 Nexusflow/Athene-V2-Agent
- 0.76
Qwen2.5-72B
- 0.75 llm-jp/llm-jp-3.1-8x13b-instruct4
- 0.75 grok-beta
- 0.74 abeja/ABEJA-Qwen2.5-32b-Japanese-v0.1
- 0.74
MiniMax-Text-01
- 0.74
Gemini 1.5 Flash
- 0.74
Mistral Large 2
- 0.74 meta-llama/Meta-Llama-3.1-405B-Instruct-FP8
- 0.74 Qwen/Qwen2.5-32B-Instruct
- 0.73 Claude 3 Opus
- 0.73
Qwen2-72B
- 0.73
Amazon Nova Pro
- 0.73 gpt-oss-20b
- 0.73 plamo-2.0-prime
- 0.73 GPT-4o mini
- 0.73 AXCXEPT/EZO-Qwen2.5-72B-Instruct
- 0.73
Gemini 1.5 Pro
- 0.72
Gemma 3 27B
- 0.72 google/gemma-2-27b-it
- 0.72
Mistral Small 3.1
- 0.72
Mistral Large 2
- 0.72 GPT-4
- 0.71 llm-jp/llm-jp-3.1-13b-instruct4
- 0.71 tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.3
- 0.71
Llama 4 Maverick
- 0.71 AXCXEPT/EZO-Qwen2.5-32B-Instruct
- 0.71
Gemini 1.5 Flash 8B
- 0.70
Llama 3.2 90B
- 0.70 plamo-1.0-prime
- 0.70 Claude 3 Sonnet
- 0.70
Gemini 1.5 Flash
- 0.70 Qwen/Qwen2.5-14B-Instruct
- 0.70 command-r-plus-08-2024
- 0.70
Llama 3.3 70B Instruct
- 0.70 cyberagent/Llama-3.1-70B-Japanese-Instruct-2407
- 0.70
Llama-3.1-Nemotron-70B-Instruct
- 0.69
Gemma 3 12B
- 0.69 meta-llama/Meta-Llama-3.1-70B-Instruct
- 0.69
Aya Expanse 32B
- 0.68 CohereForAI/c4ai-command-r-plus-08-2024
- 0.68 solar-pro-241126
- 0.68
Amazon Nova Micro
- 0.68 llm-jp/llm-jp-3-8x13b-instruct3
- 0.67 Claude 3 Haiku
- 0.67 tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.1
- 0.67 ailab-rl/model-share/model_calm3_finetuning:v0
- 0.66 llm-jp/llm-jp-3.1-1.8b-instruct4
- 0.66
Amazon Nova Lite
- 0.66 meta-llama/Meta-Llama-3-70B-Instruct
- 0.66 Qwen/Qwen2.5-7B-Instruct
- 0.66 syn-karakuri-upstage-10b
- 0.65 c4ai-command-r-08-2024
- 0.65 tokyotech-llm/Llama-3-Swallow-70B-Instruct-v0.1
- 0.65
Llama 4 Scout
- 0.65 cyberagent/calm3-22b-chat
- 0.65
Phi-4
- 0.65 solar-mini
- 0.64 llm-jp/llm-jp-3-172b-instruct3
- 0.64
Aya Expanse 8B
- 0.63
QwQ-32B-Preview
- 0.63 turing-motors/Llama-3-heron-brain-70B-v0.3
- 0.63 llm-jp/llm-jp-3-172b-beta2-instruct2
- 0.63 tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.2
- 0.62
Mistral NeMo
- 0.62 elyza/Llama-3-ELYZA-JP-8B
- 0.61 tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.1
- 1 o3 0.83
- 2 Claude Sonnet 4 Extended Thinking 0.83
- 3 o1 0.82
- 4 Claude 3.7 Sonnet thinking 0.82
- 5 Claude 3.5 Sonnet 0.82
- 6 Claude Opus 4 Extended Thinking 0.81
- 7 Claude 3.5 Sonnet (October 2024) 0.81
- 8 GPT-4.5 0.80
- 9 o1-preview 0.80
- 10 o4-mini high 0.80
- 11 chatgpt-4o-latest 0.79
- 12 GPT-4o May 2024 0.79
- 13 Grok 3 0.79
- 14
Gemini 2.5 Flash Jun 2025 0.79
- 15
Qwen2.5-Max 0.79
- 16 GPT-4.1 0.79
- 17 GPT-4o Aug 2024 0.78
- 18
Gemini 1.5 Pro Sep 2024 0.78
- 19 o3-mini high 0.78
- 20 deepseek-chat 0.78
- 21
Gemini 2.0 Flash Dec 2024 0.78
- 22 GPT-4o Nov 2024 0.77
- 23
Gemini 2.5 Pro Jun 2025 0.77
- 24 gpt-oss-120b 0.77
- 25
DeepSeek-R1 0.77
- 26
Cohere Command A 0.77
- 27
Athene-V2-Chat 0.76
- 28 GPT-4 Turbo Apr 2024 0.76
- 29 Nexusflow/Athene-V2-Agent 0.76
- 30
Qwen2.5-72B 0.76
- 31 llm-jp/llm-jp-3.1-8x13b-instruct4 0.75
- 32 grok-beta 0.75
- 33 abeja/ABEJA-Qwen2.5-32b-Japanese-v0.1 0.74
- 34
MiniMax-Text-01 0.74
- 35
Gemini 1.5 Flash Sep 2024 0.74
- 36
Mistral Large 2 Nov 2024 0.74
- 37 meta-llama/Meta-Llama-3.1-405B-Instruct-FP8 0.74
- 38 Qwen/Qwen2.5-32B-Instruct 0.74
- 39 Claude 3 Opus 0.73
- 40
Qwen2-72B 0.73
- 41
Amazon Nova Pro 0.73
- 42 gpt-oss-20b 0.73
- 43 plamo-2.0-prime 0.73
- 44 GPT-4o mini 0.73
- 45 AXCXEPT/EZO-Qwen2.5-72B-Instruct 0.73
- 46
Gemini 1.5 Pro May 2024 0.73
- 47
Gemma 3 27B 0.72
- 48 google/gemma-2-27b-it 0.72
- 49
Mistral Small 3.1 0.72
- 50
Mistral Large 2 Jul 2024 0.72
- 51 GPT-4 Jun 2023 0.72
- 52 llm-jp/llm-jp-3.1-13b-instruct4 0.71
- 53 tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.3 0.71
- 54
Llama 4 Maverick 0.71
- 55 AXCXEPT/EZO-Qwen2.5-32B-Instruct 0.71
- 56
Gemini 1.5 Flash 8B 0.71
- 57
Llama 3.2 90B 0.70
- 58 plamo-1.0-prime 0.70
- 59 Claude 3 Sonnet 0.70
- 60
Gemini 1.5 Flash May 2024 0.70
- 61 Qwen/Qwen2.5-14B-Instruct 0.70
- 62 command-r-plus-08-2024 0.70
- 63
Llama 3.3 70B Instruct 0.70
- 64 cyberagent/Llama-3.1-70B-Japanese-Instruct-2407 0.70
- 65
Llama-3.1-Nemotron-70B-Instruct 0.70
- 66
Gemma 3 12B 0.69
- 67 meta-llama/Meta-Llama-3.1-70B-Instruct 0.69
- 68
Aya Expanse 32B 0.69
- 69 CohereForAI/c4ai-command-r-plus-08-2024 0.68
- 70 solar-pro-241126 0.68
- 71
Amazon Nova Micro 0.68
- 72 llm-jp/llm-jp-3-8x13b-instruct3 0.68
- 73 Claude 3 Haiku 0.67
- 74 tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.1 0.67
- 75 ailab-rl/model-share/model_calm3_finetuning:v0 0.67
- 76 llm-jp/llm-jp-3.1-1.8b-instruct4 0.66
- 77
Amazon Nova Lite 0.66
- 78 meta-llama/Meta-Llama-3-70B-Instruct 0.66
- 79 Qwen/Qwen2.5-7B-Instruct 0.66
- 80 syn-karakuri-upstage-10b 0.66
- 81 c4ai-command-r-08-2024 0.65
- 82 tokyotech-llm/Llama-3-Swallow-70B-Instruct-v0.1 0.65
- 83
Llama 4 Scout 0.65
- 84 cyberagent/calm3-22b-chat 0.65
- 85
Phi-4 0.65
- 86 solar-mini 0.65
- 87 llm-jp/llm-jp-3-172b-instruct3 0.64
- 88
Aya Expanse 8B 0.64
- 89
QwQ-32B-Preview 0.63
- 90 turing-motors/Llama-3-heron-brain-70B-v0.3 0.63
- 91 llm-jp/llm-jp-3-172b-beta2-instruct2 0.63
- 92 tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.2 0.63
- 93
Mistral NeMo 0.62
- 94 elyza/Llama-3-ELYZA-JP-8B 0.62
- 95 tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.1 0.61
Source: Official Nejumi LLM Leaderboard 3 leaderboard
- 0.83 o3
- 0.83 Claude Sonnet 4
- 0.82 o1
- 0.82 Claude 3.7 Sonnet
- 0.82 Claude 3.5 Sonnet
- 0.81 Claude Opus 4
- 0.81 Claude 3.5 Sonnet (October 2024)
- 0.80 GPT-4.5
- 0.80 o1-preview
- 0.80 o4-mini
- 0.79 GPT-4o
- 0.79 Grok 3
- 0.79
Gemini 2.5 Flash
- 0.79
Qwen2.5-Max
- 0.79 GPT-4.1
- 0.78 GPT-4o
- 0.78
Gemini 1.5 Pro
- 0.78 o3-mini
- 0.78
Gemini 2.0 Flash
- 0.77 GPT-4o
- 0.77
Gemini 2.5 Pro
- 0.77 gpt-oss-120b
- 0.77
DeepSeek-R1
- 0.77
Cohere Command A
- 0.76
Athene-V2-Chat
- 0.76 GPT-4 Turbo
- 0.76
Qwen2.5-72B
- 0.74
MiniMax-Text-01
- 0.74
Gemini 1.5 Flash
- 0.74
Mistral Large 2
- 0.73 Claude 3 Opus
- 0.73
Qwen2-72B
- 0.73
Amazon Nova Pro
- 0.73 gpt-oss-20b
- 0.73 GPT-4o mini
- 0.73
Gemini 1.5 Pro
- 0.72
Gemma 3 27B
- 0.72
Mistral Small 3.1
- 0.72
Mistral Large 2
- 0.72 GPT-4
- 0.71
Llama 4 Maverick
- 0.71
Gemini 1.5 Flash 8B
- 0.70
Llama 3.2 90B
- 0.70 Claude 3 Sonnet
- 0.70
Gemini 1.5 Flash
- 0.70
Llama 3.3 70B Instruct
- 0.70
Llama-3.1-Nemotron-70B-Instruct
- 0.69
Gemma 3 12B
- 0.69
Aya Expanse 32B
- 0.68
Amazon Nova Micro
- 0.67 Claude 3 Haiku
- 0.66
Amazon Nova Lite
- 0.65 c4ai-command-r-08-2024
- 0.65
Llama 4 Scout
- 0.65
Phi-4
- 0.64
Aya Expanse 8B
- 0.63
QwQ-32B-Preview
- 0.62
Mistral NeMo
- 1 o3 0.83
- 2 Claude Sonnet 4 Extended Thinking 0.83
- 3 o1 0.82
- 4 Claude 3.7 Sonnet thinking 0.82
- 5 Claude 3.5 Sonnet 0.82
- 6 Claude Opus 4 Extended Thinking 0.81
- 7 Claude 3.5 Sonnet (October 2024) 0.81
- 8 GPT-4.5 0.80
- 9 o1-preview 0.80
- 10 o4-mini high 0.80
- 11 GPT-4o May 2024 0.79
- 12 Grok 3 0.79
- 13
Gemini 2.5 Flash Jun 2025 0.79
- 14
Qwen2.5-Max 0.79
- 15 GPT-4.1 0.79
- 16 GPT-4o Aug 2024 0.78
- 17
Gemini 1.5 Pro Sep 2024 0.78
- 18 o3-mini high 0.78
- 19
Gemini 2.0 Flash Dec 2024 0.78
- 20 GPT-4o Nov 2024 0.77
- 21
Gemini 2.5 Pro Jun 2025 0.77
- 22 gpt-oss-120b 0.77
- 23
DeepSeek-R1 0.77
- 24
Cohere Command A 0.77
- 25
Athene-V2-Chat 0.76
- 26 GPT-4 Turbo Apr 2024 0.76
- 27
Qwen2.5-72B 0.76
- 28
MiniMax-Text-01 0.74
- 29
Gemini 1.5 Flash Sep 2024 0.74
- 30
Mistral Large 2 Nov 2024 0.74
- 31 Claude 3 Opus 0.73
- 32
Qwen2-72B 0.73
- 33
Amazon Nova Pro 0.73
- 34 gpt-oss-20b 0.73
- 35 GPT-4o mini 0.73
- 36
Gemini 1.5 Pro May 2024 0.73
- 37
Gemma 3 27B 0.72
- 38
Mistral Small 3.1 0.72
- 39
Mistral Large 2 Jul 2024 0.72
- 40 GPT-4 Jun 2023 0.72
- 41
Llama 4 Maverick 0.71
- 42
Gemini 1.5 Flash 8B 0.71
- 43
Llama 3.2 90B 0.70
- 44 Claude 3 Sonnet 0.70
- 45
Gemini 1.5 Flash May 2024 0.70
- 46
Llama 3.3 70B Instruct 0.70
- 47
Llama-3.1-Nemotron-70B-Instruct 0.70
- 48
Gemma 3 12B 0.69
- 49
Aya Expanse 32B 0.69
- 50
Amazon Nova Micro 0.68
- 51 Claude 3 Haiku 0.67
- 52
Amazon Nova Lite 0.66
- 53 c4ai-command-r-08-2024 0.65
- 54
Llama 4 Scout 0.65
- 55
Phi-4 0.65
- 56
Aya Expanse 8B 0.64
- 57
QwQ-32B-Preview 0.63
- 58
Mistral NeMo 0.62
Source: Official Nejumi LLM Leaderboard 3 leaderboard
| # | Entry | Score | Setting | Released | Run on | Link | Source |
|---|---|---|---|---|---|---|---|
| 1 | o3 | 0.83 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 2 | Claude Sonnet 4 | 0.83 | Extended Thinking | May 2025 | Jun 2025 | wandb.ai | Official leaderboard |
| 3 | o1 | 0.82 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 4 | Claude 3.7 Sonnet | 0.82 | thinking | Feb 2025 | Apr 2025 | wandb.ai | Official leaderboard |
| 5 | Claude 3.5 Sonnet | 0.82 | Jun 2024 | Jul 2024 | wandb.ai | Official leaderboard | |
| 6 | Claude Opus 4 | 0.81 | Extended Thinking | May 2025 | Jun 2025 | wandb.ai | Official leaderboard |
| 7 | Claude 3.5 Sonnet (October 2024) | 0.81 | Oct 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 8 | GPT-4.5 | 0.80 | Feb 2025 | Mar 2025 | wandb.ai | Official leaderboard | |
| 9 | o1-preview | 0.80 | Sep 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 10 | o4-mini | 0.80 | high | Apr 2025 | Jun 2025 | wandb.ai | Official leaderboard |
| 11 | chatgpt-4o-latest | 0.79 | Aug 2024 | wandb.ai | Official leaderboard | ||
| 12 | GPT-4o May 2024 | 0.79 | May 2024 | Jul 2024 | wandb.ai | Official leaderboard | |
| 13 | Grok 3 | 0.79 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 14 | | 0.79 | Jun 2025 | Jun 2025 | wandb.ai | Official leaderboard | |
| 15 | | 0.79 | Jan 2025 | Feb 2025 | wandb.ai | Official leaderboard | |
| 16 | GPT-4.1 | 0.79 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 17 | GPT-4o Aug 2024 | 0.78 | Aug 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 18 | | 0.78 | Sep 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 19 | o3-mini | 0.78 | high | Jan 2025 | Apr 2025 | wandb.ai | Official leaderboard |
| 20 | deepseek-chat | 0.78 | Dec 2024 | wandb.ai | Official leaderboard | ||
| 21 | | 0.78 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 22 | GPT-4o Nov 2024 | 0.77 | Nov 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 23 | | 0.77 | Jun 2025 | Jun 2025 | wandb.ai | Official leaderboard | |
| 24 | gpt-oss-120b | 0.77 | Aug 2025 | Aug 2025 | wandb.ai | Official leaderboard | |
| 25 | | 0.77 | Jan 2025 | Jan 2025 | wandb.ai | Official leaderboard | |
| 26 | | 0.77 | Mar 2025 | Mar 2025 | wandb.ai | Official leaderboard | |
| 27 | | 0.76 | Nov 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 28 | GPT-4 Turbo Apr 2024 | 0.76 | Apr 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 29 | Nexusflow/Athene-V2-Agent 72.7B | 0.76 | Nov 2024 | wandb.ai | Official leaderboard | ||
| 30 | | 0.76 | Sep 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 31 | llm-jp/llm-jp-3.1-8x13b-instruct4 73.3B | 0.75 | Jun 2025 | wandb.ai | Official leaderboard | ||
| 32 | grok-beta | 0.75 | Oct 2024 | wandb.ai | Official leaderboard | ||
| 33 | abeja/ABEJA-Qwen2.5-32b-Japanese-v0.1 32.8B | 0.74 | Jan 2025 | wandb.ai | Official leaderboard | ||
| 34 | | 0.74 | Jan 2025 | Jan 2025 | wandb.ai | Official leaderboard | |
| 35 | | 0.74 | Aug 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 36 | | 0.74 | Nov 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 37 | meta-llama/Meta-Llama-3.1-405B-Instruct-FP8 405.9B | 0.74 | Sep 2024 | wandb.ai | Official leaderboard | ||
| 38 | Qwen/Qwen2.5-32B-Instruct 32.8B | 0.74 | Sep 2024 | wandb.ai | Official leaderboard | ||
| 39 | Claude 3 Opus | 0.73 | Feb 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 40 | | 0.73 | Jun 2024 | Jul 2024 | wandb.ai | Official leaderboard | |
| 41 | | 0.73 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 42 | gpt-oss-20b | 0.73 | Aug 2025 | Aug 2025 | wandb.ai | Official leaderboard | |
| 43 | plamo-2.0-prime | 0.73 | May 2025 | wandb.ai | Official leaderboard | ||
| 44 | GPT-4o mini | 0.73 | Jul 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 45 | AXCXEPT/EZO-Qwen2.5-72B-Instruct 72.7B | 0.73 | Oct 2024 | wandb.ai | Official leaderboard | ||
| 46 | | 0.73 | May 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 47 | | 0.72 | Mar 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 48 | google/gemma-2-27b-it 27.2B | 0.72 | Sep 2024 | wandb.ai | Official leaderboard | ||
| 49 | | 0.72 | Mar 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 50 | | 0.72 | Jul 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 51 | GPT-4 Jun 2023 | 0.72 | Jun 2023 | Aug 2024 | wandb.ai | Official leaderboard | |
| 52 | llm-jp/llm-jp-3.1-13b-instruct4 14B | 0.71 | Jun 2025 | wandb.ai | Official leaderboard | ||
| 53 | tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.3 70.6B | 0.71 | Dec 2024 | wandb.ai | Official leaderboard | ||
| 54 | | 0.71 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 55 | AXCXEPT/EZO-Qwen2.5-32B-Instruct 32.8B | 0.71 | Oct 2024 | wandb.ai | Official leaderboard | ||
| 56 | | 0.71 | Oct 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 57 | | 0.70 | Sep 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 58 | plamo-1.0-prime | 0.70 | Dec 2024 | wandb.ai | Official leaderboard | ||
| 59 | Claude 3 Sonnet | 0.70 | Feb 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 60 | | 0.70 | May 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 61 | Qwen/Qwen2.5-14B-Instruct 14.8B | 0.70 | Sep 2024 | wandb.ai | Official leaderboard | ||
| 62 | command-r-plus-08-2024 | 0.70 | Sep 2024 | wandb.ai | Official leaderboard | ||
| 63 | | 0.70 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 64 | cyberagent/Llama-3.1-70B-Japanese-Instruct-2407 70.6B | 0.70 | Aug 2024 | wandb.ai | Official leaderboard | ||
| 65 | | 0.70 | Oct 2024 | Oct 2024 | wandb.ai | Official leaderboard | |
| 66 | | 0.69 | Mar 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 67 | meta-llama/Meta-Llama-3.1-70B-Instruct 70.6B | 0.69 | Aug 2024 | wandb.ai | Official leaderboard | ||
| 68 | | 0.69 | Oct 2024 | Oct 2024 | wandb.ai | Official leaderboard | |
| 69 | CohereForAI/c4ai-command-r-plus-08-2024 103.8B | 0.68 | Sep 2024 | wandb.ai | Official leaderboard | ||
| 70 | solar-pro-241126 | 0.68 | Dec 2024 | wandb.ai | Official leaderboard | ||
| 71 | | 0.68 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 72 | llm-jp/llm-jp-3-8x13b-instruct3 172.4B | 0.68 | Apr 2025 | wandb.ai | Official leaderboard | ||
| 73 | Claude 3 Haiku | 0.67 | Mar 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 74 | tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.1 70.6B | 0.67 | Nov 2024 | wandb.ai | Official leaderboard | ||
| 75 | ailab-rl/model-share/model_calm3_finetuning:v0 | 0.67 | Jan 2025 | wandb.ai | Official leaderboard | ||
| 76 | llm-jp/llm-jp-3.1-1.8b-instruct4 1.8B | 0.66 | Jun 2025 | wandb.ai | Official leaderboard | ||
| 77 | | 0.66 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 78 | meta-llama/Meta-Llama-3-70B-Instruct 70.6B | 0.66 | Jul 2024 | wandb.ai | Official leaderboard | ||
| 79 | Qwen/Qwen2.5-7B-Instruct 7.6B | 0.66 | Sep 2024 | wandb.ai | Official leaderboard | ||
| 80 | syn-karakuri-upstage-10b | 0.66 | Mar 2025 | wandb.ai | Official leaderboard | ||
| 81 | c4ai-command-r-08-2024 | 0.65 | Aug 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 82 | tokyotech-llm/Llama-3-Swallow-70B-Instruct-v0.1 70.6B | 0.65 | Aug 2024 | wandb.ai | Official leaderboard | ||
| 83 | | 0.65 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 84 | cyberagent/calm3-22b-chat 22.1B | 0.65 | Jul 2024 | wandb.ai | Official leaderboard | ||
| 85 | | 0.65 | Dec 2024 | Jan 2025 | wandb.ai | Official leaderboard | |
| 86 | solar-mini | 0.65 | Feb 2025 | wandb.ai | Official leaderboard | ||
| 87 | llm-jp/llm-jp-3-172b-instruct3 172.4B | 0.64 | Apr 2025 | wandb.ai | Official leaderboard | ||
| 88 | | 0.64 | Oct 2024 | Oct 2024 | wandb.ai | Official leaderboard | |
| 89 | | 0.63 | Nov 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 90 | turing-motors/Llama-3-heron-brain-70B-v0.3 70.6B | 0.63 | Aug 2024 | wandb.ai | Official leaderboard | ||
| 91 | llm-jp/llm-jp-3-172b-beta2-instruct2 172.4B | 0.63 | Nov 2024 | wandb.ai | Official leaderboard | ||
| 92 | tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.2 8B | 0.63 | Nov 2024 | wandb.ai | Official leaderboard | ||
| 93 | | 0.62 | Jul 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 94 | elyza/Llama-3-ELYZA-JP-8B 8B | 0.62 | Jul 2024 | wandb.ai | Official leaderboard | ||
| 95 | tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.1 8B | 0.61 | Nov 2024 | wandb.ai | Official leaderboard |
| # | Model | Score | Setting | Released | Run on | Link | Source |
|---|---|---|---|---|---|---|---|
| 1 | o3 | 0.83 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 2 | Claude Sonnet 4 | 0.83 | Extended Thinking | May 2025 | Jun 2025 | wandb.ai | Official leaderboard |
| 3 | o1 | 0.82 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 4 | Claude 3.7 Sonnet | 0.82 | thinking | Feb 2025 | Apr 2025 | wandb.ai | Official leaderboard |
| 5 | Claude 3.5 Sonnet | 0.82 | Jun 2024 | Jul 2024 | wandb.ai | Official leaderboard | |
| 6 | Claude Opus 4 | 0.81 | Extended Thinking | May 2025 | Jun 2025 | wandb.ai | Official leaderboard |
| 7 | Claude 3.5 Sonnet (October 2024) | 0.81 | Oct 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 8 | GPT-4.5 | 0.80 | Feb 2025 | Mar 2025 | wandb.ai | Official leaderboard | |
| 9 | o1-preview | 0.80 | Sep 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 10 | o4-mini | 0.80 | high | Apr 2025 | Jun 2025 | wandb.ai | Official leaderboard |
| 11 | GPT-4o May 2024 | 0.79 | May 2024 | Jul 2024 | wandb.ai | Official leaderboard | |
| 12 | Grok 3 | 0.79 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 13 | | 0.79 | Jun 2025 | Jun 2025 | wandb.ai | Official leaderboard | |
| 14 | | 0.79 | Jan 2025 | Feb 2025 | wandb.ai | Official leaderboard | |
| 15 | GPT-4.1 | 0.79 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 16 | GPT-4o Aug 2024 | 0.78 | Aug 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 17 | | 0.78 | Sep 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 18 | o3-mini | 0.78 | high | Jan 2025 | Apr 2025 | wandb.ai | Official leaderboard |
| 19 | | 0.78 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 20 | GPT-4o Nov 2024 | 0.77 | Nov 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 21 | | 0.77 | Jun 2025 | Jun 2025 | wandb.ai | Official leaderboard | |
| 22 | gpt-oss-120b | 0.77 | Aug 2025 | Aug 2025 | wandb.ai | Official leaderboard | |
| 23 | | 0.77 | Jan 2025 | Jan 2025 | wandb.ai | Official leaderboard | |
| 24 | | 0.77 | Mar 2025 | Mar 2025 | wandb.ai | Official leaderboard | |
| 25 | | 0.76 | Nov 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 26 | GPT-4 Turbo Apr 2024 | 0.76 | Apr 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 27 | | 0.76 | Sep 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 28 | | 0.74 | Jan 2025 | Jan 2025 | wandb.ai | Official leaderboard | |
| 29 | | 0.74 | Aug 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 30 | | 0.74 | Nov 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 31 | Claude 3 Opus | 0.73 | Feb 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 32 | | 0.73 | Jun 2024 | Jul 2024 | wandb.ai | Official leaderboard | |
| 33 | | 0.73 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 34 | gpt-oss-20b | 0.73 | Aug 2025 | Aug 2025 | wandb.ai | Official leaderboard | |
| 35 | GPT-4o mini | 0.73 | Jul 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 36 | | 0.73 | May 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 37 | | 0.72 | Mar 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 38 | | 0.72 | Mar 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 39 | | 0.72 | Jul 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 40 | GPT-4 Jun 2023 | 0.72 | Jun 2023 | Aug 2024 | wandb.ai | Official leaderboard | |
| 41 | | 0.71 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 42 | | 0.71 | Oct 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 43 | | 0.70 | Sep 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 44 | Claude 3 Sonnet | 0.70 | Feb 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 45 | | 0.70 | May 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 46 | | 0.70 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 47 | | 0.70 | Oct 2024 | Oct 2024 | wandb.ai | Official leaderboard | |
| 48 | | 0.69 | Mar 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 49 | | 0.69 | Oct 2024 | Oct 2024 | wandb.ai | Official leaderboard | |
| 50 | | 0.68 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 51 | Claude 3 Haiku | 0.67 | Mar 2024 | Aug 2024 | wandb.ai | Official leaderboard | |
| 52 | | 0.66 | Dec 2024 | Dec 2024 | wandb.ai | Official leaderboard | |
| 53 | c4ai-command-r-08-2024 | 0.65 | Aug 2024 | Sep 2024 | wandb.ai | Official leaderboard | |
| 54 | | 0.65 | Apr 2025 | Apr 2025 | wandb.ai | Official leaderboard | |
| 55 | | 0.65 | Dec 2024 | Jan 2025 | wandb.ai | Official leaderboard | |
| 56 | | 0.64 | Oct 2024 | Oct 2024 | wandb.ai | Official leaderboard | |
| 57 | | 0.63 | Nov 2024 | Nov 2024 | wandb.ai | Official leaderboard | |
| 58 | | 0.62 | Jul 2024 | Aug 2024 | wandb.ai | Official leaderboard |
Rankings by part
We read 22 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as 汎用的言語性能(GLP)_AVG, アラインメント(ALT)_AVG and GLP_表現. o3 leads 7 of them, and 8 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
汎用的言語性能(GLP)_AVG
58 models we track have a score here.
- 1 o3 0.81
- 2 o1 ↑1 0.80
- 3 o1-preview ↑6 0.79
- 4 o4-mini ↑6 0.78
- 5 Claude 3.7 Sonnet ↓1 0.78
- 6 GPT-4.1 ↑9 0.78
- 7 Claude Opus 4 ↓1 0.78
- 8 Grok 3 ↑4 0.77
- 9 o3-mini ↑9 0.77
- 10 Claude Sonnet 4 ↓8 0.77
アラインメント(ALT)_AVG
58 models we track have a score here.
- 1 Claude Sonnet 4 ↑1 0.88
- 2 Claude 3.5 Sonnet ↑3 0.86
- 3 Claude 3.7 Sonnet ↑1 0.86
- 4 o3 ↓3 0.86
- 5 Claude 3.5 Sonnet (October 2024) ↑2 0.86
- 6 o1 ↓3 0.85
- 7 Claude Opus 4 ↓1 0.85
- 8 GPT-4o May 2024 ↑3 0.85
- 9 GPT-4.5 ↓1 0.84
- 10
Gemini 1.5 Pro Sep 2024 ↑7 0.84
GLP_表現
58 models we track have a score here.
- 1 Claude Sonnet 4 ↑1 0.95
- 2
Gemini 2.5 Pro Jun 2025 ↑19 0.95
- 3 Claude Opus 4 ↑3 0.93
- 4
Gemini 2.5 Flash Jun 2025 ↑9 0.93
- 5 o1 ↓2 0.92
- 6 Grok 3 ↑6 0.92
- 7 Claude 3.5 Sonnet (October 2024) 0.91
- 8
DeepSeek-R1 ↑15 0.91
- 9 o3 ↓8 0.91
- 10 o3-mini ↑8 0.91
GLP_翻訳
58 models we track have a score here.
- 1 GPT-4.5 ↑7 0.88
- 2 Claude Opus 4 ↑4 0.87
- 3
Gemini 2.0 Flash Dec 2024 ↑16 0.87
- 4 Claude 3 Opus ↑27 0.87
- 5 GPT-4o Nov 2024 ↑15 0.87
- 6
Cohere Command A ↑18 0.87
- 7 GPT-4.1 ↑8 0.87
- 8 Claude 3.7 Sonnet ↓4 0.87
- 9 Claude 3.5 Sonnet (October 2024) ↓2 0.87
- 10 GPT-4o May 2024 ↑1 0.87
GLP_情報検索
58 models we track have a score here.
- 1 GPT-4 Jun 2023 ↑39 0.90
- 2
Amazon Nova Pro ↑31 0.90
- 3
Amazon Nova Lite ↑49 0.89
- 4
Qwen2.5-Max ↑10 0.89
- 5
Qwen2.5-72B ↑22 0.88
- 6
Phi-4 ↑49 0.88
- 7
Athene-V2-Chat ↑18 0.88
- 8
MiniMax-Text-01 ↑20 0.88
- 9
Llama 3.2 90B ↑34 0.88
- 10 GPT-4o mini ↑25 0.88
GLP_推論
58 models we track have a score here.
- 1 Claude 3.5 Sonnet ↑4 0.83
- 2 Grok 3 ↑10 0.82
- 3 Claude 3.7 Sonnet ↑1 0.82
- 4 GPT-4.1 ↑11 0.81
- 5
Athene-V2-Chat ↑20 0.80
- 6
Llama-3.1-Nemotron-70B-Instruct ↑41 0.79
- 7
DeepSeek-R1 ↑16 0.77
- 8 Claude Sonnet 4 ↓6 0.77
- 9 o1-preview 0.76
- 10 gpt-oss-120b ↑12 0.76
GLP_数学的推論
58 models we track have a score here.
- 1 o1 ↑2 0.93
- 2 o3-mini ↑16 0.93
- 3 Claude Opus 4 ↑3 0.91
- 4 o1-preview ↑5 0.91
- 5 o3 ↓4 0.91
- 6
Gemini 2.5 Flash Jun 2025 ↑7 0.91
- 7 Claude Sonnet 4 ↓5 0.91
- 8 o4-mini ↑2 0.90
- 9
Gemini 2.5 Pro Jun 2025 ↑12 0.90
- 10 gpt-oss-20b ↑24 0.90
GLP_抽出
58 models we track have a score here.
- 1 Claude Sonnet 4 ↑1 0.40
- 2 Claude Opus 4 ↑4 0.39
- 3 GPT-4.5 ↑5 0.39
- 4
DeepSeek-R1 ↑19 0.38
- 5 GPT-4o Aug 2024 ↑11 0.38
- 6 GPT-4o Nov 2024 ↑14 0.38
- 7 Claude 3 Opus ↑24 0.38
- 8 GPT-4.1 ↑7 0.38
- 9 o1-preview 0.38
- 10 GPT-4 Turbo Apr 2024 ↑16 0.38
GLP_知識・質問応答
58 models we track have a score here.
- 1 o3 0.90
- 2 Claude Opus 4 ↑4 0.88
- 3 o1 0.88
- 4 o1-preview ↑5 0.88
- 5 Claude Sonnet 4 ↓3 0.87
- 6 GPT-4.5 ↑2 0.86
- 7 o4-mini ↑3 0.85
- 8
Gemini 2.5 Flash Jun 2025 ↑5 0.85
- 9
Gemini 2.5 Pro Jun 2025 ↑12 0.85
- 10 GPT-4o Nov 2024 ↑10 0.83
GLP_英語
58 models we track have a score here.
- 1 o3 0.95
- 2 o1 ↑1 0.94
- 3 o1-preview ↑6 0.93
- 4 o4-mini ↑6 0.93
- 5 gpt-oss-120b ↑17 0.90
- 6 o3-mini ↑12 0.90
- 7 GPT-4.5 ↑1 0.88
- 8
Gemini 1.5 Pro Sep 2024 ↑9 0.87
- 9 Grok 3 ↑3 0.86
- 10 Claude 3.7 Sonnet ↓6 0.86
GLP_意味解析
58 models we track have a score here.
- 1 o3 0.86
- 2 o1 ↑1 0.85
- 3 o4-mini ↑7 0.85
- 4 Claude Opus 4 ↑2 0.83
- 5
Qwen2.5-Max ↑9 0.83
- 6 Claude Sonnet 4 ↓4 0.83
- 7 GPT-4.1 ↑8 0.83
- 8 GPT-4.5 0.82
- 9 o3-mini ↑9 0.82
- 10 o1-preview ↓1 0.81
GLP_構文解析
58 models we track have a score here.
- 1 GPT-4.5 ↑7 0.71
- 2 Claude Opus 4 ↑4 0.71
- 3 o3 ↓2 0.69
- 4 o1 ↓1 0.68
- 5 Claude 3.5 Sonnet 0.67
- 6 Claude Sonnet 4 ↓4 0.67
- 7 Claude 3.7 Sonnet ↓3 0.67
- 8 GPT-4 Jun 2023 ↑32 0.66
- 9 o1-preview 0.66
- 10 Claude 3.5 Sonnet (October 2024) ↓3 0.66
ALT_制御性
58 models we track have a score here.
- 1 o1 ↑2 0.97
- 2 o3 ↓1 0.96
- 3 o3-mini ↑15 0.96
- 4 o4-mini ↑6 0.96
- 5 gpt-oss-120b ↑17 0.94
- 6 GPT-4.5 ↑2 0.93
- 7 Claude Opus 4 ↓1 0.93
- 8 Claude 3.5 Sonnet (October 2024) ↓1 0.93
- 9 Claude 3.5 Sonnet ↓4 0.92
- 10 Claude Sonnet 4 ↓8 0.92
ALT_倫理・道徳
58 models we track have a score here.
- 1 Claude Opus 4 ↑5 0.98
- 2 Claude 3 Opus ↑29 0.97
- 3 GPT-4o mini ↑32 0.97
- 4 GPT-4o Aug 2024 ↑12 0.97
- 5 Claude 3.5 Sonnet (October 2024) ↑2 0.97
- 6
Qwen2.5-Max ↑8 0.97
- 7 Claude 3.7 Sonnet ↓3 0.97
- 8
Gemini 2.5 Pro Jun 2025 ↑13 0.97
- 9 Claude 3.5 Sonnet ↓4 0.96
- 10
Mistral Large 2 Nov 2024 ↑20 0.96
ALT_毒性
58 models we track have a score here.
- 1
Gemini 1.5 Pro May 2024 ↑35 0.88
- 2
Amazon Nova Pro ↑31 0.86
- 3 Claude 3 Sonnet ↑41 0.85
- 4 Claude 3 Opus ↑27 0.85
- 5 Claude Sonnet 4 ↓3 0.84
- 6
Gemini 1.5 Flash May 2024 ↑39 0.84
- 7 Claude 3.5 Sonnet ↓2 0.83
- 8 GPT-4o May 2024 ↑3 0.83
- 9 Claude 3.7 Sonnet ↓5 0.83
- 10 GPT-4 Turbo Apr 2024 ↑16 0.83
ALT_バイアス
58 models we track have a score here.
- 1 Claude 3.5 Sonnet ↑4 1.00
- 2 GPT-4 Turbo Apr 2024 ↑24 0.97
- 3 GPT-4o May 2024 ↑8 0.97
- 4 Claude 3.5 Sonnet (October 2024) ↑3 0.96
- 5 Claude 3.7 Sonnet ↓1 0.95
- 6 GPT-4.5 ↑2 0.95
- 7 GPT-4o Aug 2024 ↑9 0.94
- 8 o1 ↓5 0.93
- 9
DeepSeek-R1 ↑14 0.92
- 10 Claude Sonnet 4 ↓8 0.91
ALT_堅牢性
58 models we track have a score here.
- 1 o3 0.94
- 2 o1 ↑1 0.94
- 3 o3-mini ↑15 0.92
- 4 o4-mini ↑6 0.91
- 5 Claude Sonnet 4 ↓3 0.91
- 6 Claude Opus 4 0.90
- 7 o1-preview ↑2 0.89
- 8
Gemini 2.5 Flash Jun 2025 ↑5 0.85
- 9
Qwen2.5-Max ↑5 0.85
- 10
Gemini 1.5 Pro Sep 2024 ↑7 0.84
ALT_真実性
58 models we track have a score here.
- 1 o3 0.77
- 2 Claude Opus 4 ↑4 0.76
- 3 Claude Sonnet 4 ↓1 0.76
- 4
Gemini 2.5 Flash Jun 2025 ↑9 0.74
- 5
Gemini 2.0 Flash Dec 2024 ↑14 0.73
- 6 GPT-4.5 ↑2 0.73
- 7
Gemini 1.5 Pro Sep 2024 ↑10 0.72
- 8
DeepSeek-R1 ↑15 0.72
- 9
Gemini 2.5 Pro Jun 2025 ↑12 0.71
- 10 o1 ↓7 0.71
AVG_jaster_0shot
58 models we track have a score here.
- 1 o3 0.76
- 2 o1 ↑1 0.75
- 3 o1-preview ↑6 0.74
- 4 GPT-4.5 ↑4 0.74
- 5 Claude Opus 4 ↑1 0.74
- 6 o4-mini ↑4 0.73
- 7 Claude Sonnet 4 ↓5 0.72
- 8 o3-mini ↑10 0.71
- 9 Claude 3.7 Sonnet ↓5 0.71
- 10 GPT-4.1 ↑5 0.71
AVG_jaster_2shots
58 models we track have a score here.
- 1 Claude Opus 4 ↑5 0.79
- 2 o3 ↓1 0.78
- 3 o1 0.77
- 4 Claude Sonnet 4 ↓2 0.77
- 5 GPT-4.5 ↑3 0.77
- 6 o1-preview ↑3 0.76
- 7 Claude 3.7 Sonnet ↓3 0.76
- 8 o4-mini ↑2 0.75
- 9 GPT-4.1 ↑6 0.75
- 10 Claude 3.5 Sonnet (October 2024) ↓3 0.75
AVG_mtbench
58 models we track have a score here.
- 1 Grok 3 ↑11 8.91
- 2 GPT-4o Nov 2024 ↑18 8.86
- 3 Claude Sonnet 4 ↓1 8.83
- 4 GPT-4.1 ↑11 8.82
- 5 o3 ↓4 8.82
- 6 o1 ↓3 8.81
- 7 Claude Opus 4 ↓1 8.76
- 8
Qwen2.5-Max ↑6 8.74
- 9
Gemini 2.5 Flash Jun 2025 ↑4 8.70
- 10 Claude 3.5 Sonnet ↓5 8.69
AVG_lctg
58 models we track have a score here.
- 1 o1 ↑2 0.99
- 2 o3 ↓1 0.99
- 3 o3-mini ↑15 0.98
- 4 o4-mini ↑6 0.98
- 5 gpt-oss-120b ↑17 0.94
- 6 Claude Opus 4 0.94
- 7 GPT-4.5 ↑1 0.93
- 8 Claude Sonnet 4 ↓6 0.92
- 9 Claude 3.5 Sonnet (October 2024) ↓2 0.91
- 10 Claude 3.5 Sonnet ↓5 0.91
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 Nejumi LLM Leaderboard 3 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: Official Nejumi LLM Leaderboard 3 leaderboard
3 models have held number one on the official leaderboard since its release in July 2024
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Nejumi LLM Leaderboard 3 leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Nejumi LLM Leaderboard 3 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
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 (95 entries).
- Pass Models scored 58 models we track, plus 37 other entries on its leaderboard.
- Note Newest score The newest came 15 months ago, in August 2025, for gpt-oss-120b. That’s expected for a test replaced by Nejumi LLM Leaderboard 4.
- 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 Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on wandb.ai: 61 rows for models we track and 37 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.