Skip to content

Benchmark

NYT Connections – Newest 100 puzzles, extended version (Lech Mazur)

Scores are percentages.

Models measured
106
Scores recorded
138
Sources
1
IQ weight
2.05

Score distribution

61.5%
Mean
70.9%
Median
30%
Std dev
14.6%
10th percentile
94.5%
90th percentile
5.5%
Min
5.5%99.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.05 for chat models, measured over 106 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 4 Carbonhigh reasoning· Lech Mazur suite data99.6%
2Gemini 4 Neonhigh reasoning· Lech Mazur suite data99.6%
3Gemini 4 Argonhigh reasoning· Lech Mazur suite data98.7%
4GPT-6 Astraxhigh reasoning· Lech Mazur suite data98.7%
5GPT-6.1 Solmax reasoning· Lech Mazur suite data98.7%
6GPT-6.2 Astrahigh reasoning· Lech Mazur suite data98.2%
7GPT-6 Astramax reasoning· Lech Mazur suite data97.9%
8GPT-6 Astrahigh reasoning· Lech Mazur suite data97.8%
9Claude Fable 5.2high reasoning· Lech Mazur suite data96.9%
10Gemini 3.8 Flashhigh reasoning· Lech Mazur suite data96.5%
11Gemini 3.1 Pro Preview· Lech Mazur suite data96.1%
12GPT 5.5xhigh reasoning· Lech Mazur suite data96.1%
13GPT-6.1 Solhigh reasoning· Lech Mazur suite data95.6%
14GPT 5.5high reasoning· Lech Mazur suite data95.2%
15Claude Opus 5xhigh reasoning· Lech Mazur suite data94.5%
16GPT-5.6 Solxhigh reasoning· Lech Mazur suite data94.4%
17Claude Fable 5high reasoning· Lech Mazur suite data93.6%
18GPT 5.4xhigh reasoning· Lech Mazur suite data93.4%
19GPT-6.2 Solhigh reasoning· Lech Mazur suite data92.9%
20Muse Wonder 1.0high reasoning· Lech Mazur suite data92.8%
21GPT-5.6 Solhigh reasoning· Lech Mazur suite data92.7%
22Gemini 3 Pro Preview· Lech Mazur suite data92.3%
23Gemini 3.7 Flash· Lech Mazur suite data92.2%
24Claude Opus 4.8high reasoning· Lech Mazur suite data92.1%
25Kimi K3· Lech Mazur suite data92.1%
26Claude Opus 5high reasoning· Lech Mazur suite data91.9%
27DeepSeek V4.1 Flash 1017high reasoning· Lech Mazur suite data91.3%
28Gemini 3.5 Flash· Lech Mazur suite data90.8%
29Claude Opus 4.8xhigh reasoning· Lech Mazur suite data90.6%
30Claude Opus 5.5high reasoning· Lech Mazur suite data90.6%
31GPT-6.1 Terrahigh reasoning· Lech Mazur suite data90.2%
32Kimi K3.1 Prohigh reasoning· Lech Mazur suite data90.2%
33Qwen3.8 Max· Lech Mazur suite data88.9%
34GPT-6 Solhigh reasoning· Lech Mazur suite data88.8%
35Claude Fable 5.1high reasoning· Lech Mazur suite data88.5%
36Gemini 3.6 Flash· Lech Mazur suite data88.5%
37DeepSeek V4 Prohigh reasoning· Lech Mazur suite data88.4%
38GPT 5.4high reasoning· Lech Mazur suite data88.4%
39Claude Opus 4.6high reasoning· Lech Mazur suite data88.1%
40GPT 5.5medium reasoning· Lech Mazur suite data87.3%
41DeepSeek V4 Flash· Lech Mazur suite data87.2%
42Qwen4 Maxhigh· Lech Mazur suite data86.6%
43DeepSeek V4.1 Flashhigh reasoning· Lech Mazur suite data86.1%
44Hy 4high· Lech Mazur suite data86%
45Muse Spark 1.1high reasoning· Lech Mazur suite data84.8%
46GPT 5.4medium reasoning· Lech Mazur suite data84.4%
47Muse Spark 1.3high reasoning· Lech Mazur suite data83.6%
48Claude Opus 5.5medium reasoning· Lech Mazur suite data83.4%
49Kimi K2.6· Lech Mazur suite data80.2%
50Claude Sonnet 5.5high reasoning· Lech Mazur suite data79.1%

Showing 1–50 of 138 results.