NYT Connections Extended — leaderboard

Extended version: 940 puzzles with extra trick words added that don't fit any category. Models get one attempt. Significantly harder than the original.

Metric: Score (%). Source: github.com. Status: saturated. 92 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)97.8
2GPT-5.5 (xHigh)97.8
3GPT-5.5 (High)97.2
4GPT-5.4 (xHigh)96
5Gemini 3 Pro (Preview)95.2
6Gemini 3.5 Flash94.2
7GPT-5.4 (High)93
8GPT-5.5 (Medium)91.8
9Claude Opus 4.6 (Thinking, High)91.5
10Kimi K2.686.5
11Grok 4.20 0309 (Reasoning)85.5
12GPT-5.2 (xHigh)85
13Qwen 3.7 Max84.8
14GLM-5.184.2
15Grok 4.1 Fast (Reasoning)83.8

Interactive version: theaggregate.ai/benchmark?slug=nyt-connections-extended · How the rankings work · Data refreshed daily, snapshot 2026-07-22.