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
1Claude Opus 4.616k· Lech Mazur suite data85.2%
2Claude Opus 4.8xhigh· Lech Mazur suite data50%
3Muse Spark 1.1high· Lech Mazur suite data49.5%
4Gemini 4 Argonhigh· Lech Mazur suite data45.4%
5Muse Spark 1.3high· Lech Mazur suite data23.4%
6GLM 5.2max· Lech Mazur suite data18.2%
7DeepSeek V4 Prohigh· Lech Mazur suite data14%
8GPT 5medium· Epoch AI Benchmarking Hub data8.6%
9Kimi K2 Instruct· Epoch AI Benchmarking Hub data8.6%
10Claude Opus 4.1· Epoch AI Benchmarking Hub data8.5%
11Claude Opus 4.116k· Epoch AI Benchmarking Hub data8.4%
12O3 Promedium· Epoch AI Benchmarking Hub data8.4%
13OpenAI o3medium· Epoch AI Benchmarking Hub data8.4%
14Gemini 2.5 Pro· Epoch AI Benchmarking Hub data8.4%
15Claude Opus 416k· Epoch AI Benchmarking Hub data8.4%
16Claude Opus 4· Epoch AI Benchmarking Hub data8.3%
17GPT-5 minimedium· Epoch AI Benchmarking Hub data8.3%
18DeepSeek R1· Epoch AI Benchmarking Hub data8.3%
19Qwen 3 235B A22B· Epoch AI Benchmarking Hub data8.3%
20Qwen 3 235B A22B Thinking 2507· Epoch AI Benchmarking Hub data8.2%
21GPT 4o· Epoch AI Benchmarking Hub data8.2%
22Claude Sonnet 416k· Epoch AI Benchmarking Hub data8.1%
23Claude Sonnet 3.716k· Epoch AI Benchmarking Hub data8.1%
24Claude Sonnet 4· Epoch AI Benchmarking Hub data8.1%
25Gemini 2.5 Pro Preview 05-06· Epoch AI Benchmarking Hub data8.1%
26Gemini 2.5 Pro· Epoch AI Benchmarking Hub data8.1%
27Claude 3.5 Sonnet· Epoch AI Benchmarking Hub data8%
28QwQ 32B16k· Epoch AI Benchmarking Hub data8%
29Gemma 3 27B IT· Epoch AI Benchmarking Hub data8%
30Claude Sonnet 3.7· Epoch AI Benchmarking Hub data7.9%
31Mistral Medium 3· Epoch AI Benchmarking Hub data7.7%
32GPT OSS 120B· Epoch AI Benchmarking Hub data
Values from 2 sources
7.7%
33DeepSeek V3· Epoch AI Benchmarking Hub data7.7%
34Grok 4· Epoch AI Benchmarking Hub data7.7%
35gemini-2.5-flash-preview-04-1724k· Epoch AI Benchmarking Hub data7.7%
36Grok 3beta· Epoch AI Benchmarking Hub data7.6%
37GPT-4.5 Preview· Epoch AI Benchmarking Hub data7.6%
38Qwen3 30B A3B· Epoch AI Benchmarking Hub data7.5%
39o4 Minimedium· Epoch AI Benchmarking Hub data7.5%
40Gemini 2.0 Flash Experimentalthinking· Epoch AI Benchmarking Hub data7.4%
41Claude Haiku 3.5· Epoch AI Benchmarking Hub data7.4%
42Grok 3 Minilow, beta· Epoch AI Benchmarking Hub data7.4%
43GLM 4.5· Epoch AI Benchmarking Hub data7.3%
44Qwen2.5-Max· Epoch AI Benchmarking Hub data7.3%
45Gemini 2.0 Flash Experimental· Epoch AI Benchmarking Hub data7.2%
46o1medium· Epoch AI Benchmarking Hub data7%
47Mistral Large2407· Epoch AI Benchmarking Hub data6.9%
48GPT-4o mini· Epoch AI Benchmarking Hub data6.7%
49O1 Minimedium· Epoch AI Benchmarking Hub data6.5%
50Grok 2· Epoch AI Benchmarking Hub data6.4%

Showing 1–50 of 112 results.

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