AI vs Puzzles
View Puzzles
Back to Home
Puzzle
Connections Leaderboard
The leaderboard is a ranking of the Large Language Models against Connections puzzles.
Position
Model
Provider
Average Score
Correct Connections
Incorrect Connections
Completed
1
qwen/qwen3-235b-a22b
2.1
2035
1873
977
2
x-ai/grok-3-mini-beta
1.8
1910
2342
1063
3
deepseek/deepseek-chat-v3-0324
1.4
1606
3154
1190
4
anthropic/claude-3.7-sonnet
1.3
1421
2811
1058
5
openai/gpt-4o
1.2
1425
3335
1190
6
mistralai/mistral-medium-3
1.2
1475
3281
1189
7
google/gemini-2.0-flash-001
1.1
1160
3160
1080
8
meta-llama/llama-3.3-70b-instruct
1.1
1274
3486
1190