Skip to content

Benchmark

Nejumi 4: GLP_基礎的言語性能

Higher scores are better.

Models measured
114
Scores recorded
126
Sources
1
IQ weight
1.88

Score distribution

0.768
Mean
0.809
Median
0.111
Std dev
0.597
10th percentile
0.861
90th percentile
0.409
Min
0.4090.878
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.88 for chat models, measured over 114 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT 5.5openai/gpt-5.5-2026-04-23: xhigh-effort· Nejumi Leaderboard 4 data docs0.878
2GPT 5.5openai/gpt-5.5-2026-04-23: high-effort· Nejumi Leaderboard 4 data docs0.878
3Gemini 3 Flashgoogle/gemini-3-flash-preview· Nejumi Leaderboard 4 data docs0.876
4GPT-5.6 Solopenai/gpt-5.6-sol: max-effort· Nejumi Leaderboard 4 data docs0.87
5GPT-5.6 Terraopenai/gpt-5.6-terra: max-effort· Nejumi Leaderboard 4 data docs0.869
6Gemma 4 31B ITgoogle/gemma-4-31b-it· Nejumi Leaderboard 4 data docs0.868
7Claude Fable 5anthropic/claude-fable-5: adaptive-thinking-max with fallback to Opus 4.8· Nejumi Leaderboard 4 data docs0.867
8Claude Opus 4.6anthropic/claude-opus-4-6: extended-thinking· Nejumi Leaderboard 4 data docs0.867
9GPT 5.1openai/gpt-5.1-2025-11-13: high-effort· Nejumi Leaderboard 4 data docs0.863
10Gemini 3 Pro Previewgoogle/gemini-3-pro-preview· Nejumi Leaderboard 4 data docs0.862
11Gemini 3.5 Flashgoogle/gemini-3.5-flash· Nejumi Leaderboard 4 data docs0.862
12Claude Sonnet 4.6anthropic/claude-sonnet-4.6: extended-thinking· Nejumi Leaderboard 4 data docs0.861
13Gemini 3.6 Flashgoogle/gemini-3.6-flash· Nejumi Leaderboard 4 data docs0.861
14Gemini 2.5 Progoogle/gemini-2.5-pro· Nejumi Leaderboard 4 data docs0.859
15GPT-5.6 Lunaopenai/gpt-5.6-luna: max-effort· Nejumi Leaderboard 4 data docs0.859
16GPT 5.4openai/gpt-5.4-2026-03-05: high-effort· Nejumi Leaderboard 4 data docs0.859
17Claude Opus 4.8anthropic/claude-opus-4.8: adaptive-thinking-xhigh· Nejumi Leaderboard 4 data docs0.859
18Claude Opus 4.5anthropic/claude-opus-4.5-20251125: extended-thinking· Nejumi Leaderboard 4 data docs0.858
19Qwen3.5 397B A17BQwen/Qwen3.5-397B-A17B: reasoning-enabled· Nejumi Leaderboard 4 data docs0.858
20OpenAI o3openai/o3-2025-04-16: high-effort· Nejumi Leaderboard 4 data docs0.858
21Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview· Nejumi Leaderboard 4 data docs0.857
22Claude Opus 5anthropic/claude-opus-5: adaptive-thinking-max· Nejumi Leaderboard 4 data docs0.856
23GPT 5.4openai/gpt-5.4-2026-03-05: xhigh-effort· Nejumi Leaderboard 4 data docs0.856
24Qwen3.6 Max Previewqwen/qwen3.6-max-preview: openrouter-reasoning· Nejumi Leaderboard 4 data docs0.856
25Kimi K3moonshotai/kimi-k3: reasoning-max· Nejumi Leaderboard 4 data docs0.855
26Claude Opus 4.7anthropic/claude-opus-4.7: adaptive-thinking-xhigh· Nejumi Leaderboard 4 data docs0.852
27GPT 5.2openai/gpt-5.2-2025-12-11: xhigh-effort· Nejumi Leaderboard 4 data docs0.851
28Grok 4.5xai/grok-4.5: reasoning-high· Nejumi Leaderboard 4 data docs0.85
29o1openai/o1-2024-12-17: high-effort· Nejumi Leaderboard 4 data docs0.85
30Qwen3.5 122B-A10BQwen/Qwen3.5-122B-A10B: reasoning-enabled· Nejumi Leaderboard 4 data docs0.846
31GPT 5openai/gpt-5-2025-08-07: high-effort· Nejumi Leaderboard 4 data docs0.843
32DeepSeek V4 Flashdeepseek-ai/DeepSeek-V4-Flash: reasoning-max· Nejumi Leaderboard 4 data docs0.839
33Gemma 4 26B A4Bgoogle/gemma-4-26B-A4B-it· Nejumi Leaderboard 4 data docs0.839
34GPT-5.4 miniopenai/gpt-5.4-mini-2026-03-17: high-effort· Nejumi Leaderboard 4 data docs0.838
35MiniMax M3MiniMaxAI/MiniMax-M3: reasoning-enabled· Nejumi Leaderboard 4 data docs0.838
36DeepSeek R1deepseek-ai/DeepSeek-R1-0528: reasoning-enabled· Nejumi Leaderboard 4 data docs0.838
37GLM 4.7zai-org/GLM-4.7: reasoning-enabled· Nejumi Leaderboard 4 data docs0.834
38Qwen3.5 27BQwen/Qwen3.5-27B: reasoning-enabled· Nejumi Leaderboard 4 data docs0.834
39GLM 5zai-org/GLM-5: reasoning-enabled· Nejumi Leaderboard 4 data docs0.833
40Grok 4.1 Fastx-ai/grok-4-1-fast-reasoning· Nejumi Leaderboard 4 data docs0.833
41GPT 4.1openai/gpt-4-1-2025-04-14· Nejumi Leaderboard 4 data docs0.833
42Claude Sonnet 4.5anthropic/claude-sonnet-4-5-20250929: extended-thinking· Nejumi Leaderboard 4 data docs0.832
43DeepSeek Reasonerdeepseek/DeepSeek-V3.2 (Thinking Mode)· Nejumi Leaderboard 4 data docs0.831
44GLM 5.2zai-org/GLM-5.2: reasoning-max· Nejumi Leaderboard 4 data docs0.831
45Grok 4.20xai/grok-4.20-0309-reasoning· Nejumi Leaderboard 4 data docs0.829
46o4 Miniopenai/o4-mini-2025-04-16· Nejumi Leaderboard 4 data docs0.827
47GLM 4.6zai-org/GLM-4.6-FP8: reasoning-enabled· Nejumi Leaderboard 4 data docs0.827
48DeepSeek V3.2deepseek-ai/DeepSeek-V3.2: reasoning-enabled· Nejumi Leaderboard 4 data docs0.825
49Qwen3.6 27BQwen/Qwen3.6-27B: reasoning-enabled· Nejumi Leaderboard 4 data docs0.824
50Qwen3.5 35B A3BQwen/Qwen3.5-35B-A3B: reasoning-enabled· Nejumi Leaderboard 4 data docs0.822

Showing 1–50 of 126 results.

Data credits: Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results