Skip to content

Benchmark

Lech Mazur Writing

Scores are percentages.

Models measured
102
Scores recorded
112
Sources
2
IQ weight
0.00

Score distribution

5.8%
Mean
4.3%
Median
11.4%
Std dev
-2.1%
10th percentile
8.4%
90th percentile
-5.4%
Min
-5.4%85.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.00 for chat models, measured over 102 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51Phi 4· Epoch AI Benchmarking Hub data6.3%
52Llama 4 Maverick· Epoch AI Benchmarking Hub data6.2%
53o3 Minihigh· Epoch AI Benchmarking Hub data6.2%
54o3 Minimedium· Epoch AI Benchmarking Hub data6.2%
55Nova Prov1· Epoch AI Benchmarking Hub data6.1%
56GPT 5.2medium· Lech Mazur suite data5.6%
57GPT-6.2 Astrahigh· Lech Mazur suite data4.3%
58Claude Fable 5.2high· Lech Mazur suite data3.6%
59Qwen4 Maxhigh· Lech Mazur suite data3.5%
60Claude Fable 5.1high· Lech Mazur suite data3.5%
61Claude Opus 5xhigh· Lech Mazur suite data3.4%
62Claude Opus 5.5high· Lech Mazur suite data3.4%
63GPT-6 Astrahigh· Lech Mazur suite data3.2%
64Claude Opus 5high· Lech Mazur suite data3%
65Grok 4.7high· Lech Mazur suite data2.9%
66GLM 5.3max· Lech Mazur suite data2.6%
67Claude Fable 5high· Lech Mazur suite data2.6%
68Kimi K3.1 Prohigh· Lech Mazur suite data2.4%
69GPT-5.6 Solxhigh· Lech Mazur suite data2.2%
70GPT 5.5xhigh· Lech Mazur suite data2.2%
71Kimi K3· Lech Mazur suite data2.1%
72Hy 4high· Lech Mazur suite data2%
73GPT-5.6 Solhigh· Lech Mazur suite data2%
74GPT 5.4medium· Lech Mazur suite data1.7%
75GPT 5.4xhigh· Lech Mazur suite data1.7%
76Gemini 4 Carbonhigh· Lech Mazur suite data1.6%
77Muse Wonderhigh· Lech Mazur suite data1.6%
78Claude Opus 4.7high· Lech Mazur suite data1.5%
79Claude Sonnet 4.616k· Lech Mazur suite data1.4%
80MiMo V2.6 Prothinking· Lech Mazur suite data1.2%
81Claude Opus 4.8high· Lech Mazur suite data-0.013%
82Gemini 3.8 Flashhigh· Lech Mazur suite data-0.157%
83Qwen3.8 Max· Lech Mazur suite data-0.174%
84Step 5 Previewhigh· Lech Mazur suite data-0.185%
85Kimi K2.6· Lech Mazur suite data-0.191%
86Muse Spark 1.2high· Lech Mazur suite data-0.33%
87MiniMax M3· Lech Mazur suite data-0.389%
88Mistral Medium 3.1· Lech Mazur suite data-0.66%
89DeepSeek V4.1 Flashhigh· Lech Mazur suite data-0.797%
90DeepSeek-V4-Pro Preview· Lech Mazur suite data-0.814%
91Qwen 3.8 27B· Lech Mazur suite data-0.956%
92MiMo V2.5 Pro· Lech Mazur suite data-0.957%
93Qwen3 Max Preview· Lech Mazur suite data-0.959%
94Gemini 3.7 Flashhigh· Lech Mazur suite data-0.995%
95Qwen3.6 Max Preview· Lech Mazur suite data-1.2%
96GLM 5.1· Lech Mazur suite data-1.3%
97Kimi K2.5thinking· Lech Mazur suite data-1.4%
98MiMo V2 Pro· Lech Mazur suite data-1.5%
99ERNIE 5.1· Lech Mazur suite data-1.5%
100Mistral Large 4high· Lech Mazur suite data-1.9%

Showing 51–100 of 112 results.

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)