Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Puzzle Solving on PUZZLE (Human easy)
Loading...
69.5
Accuracy
GPT-5
43.812
50.481
57.15
63.819
Jun 2, 2026
Accuracy
Mean Completion Tokens
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Mean Completion Tokens
GPT-5
2026.06
69.5
10,179.6
Kimi K2
2026.06
56.2
29,713.7
DeepSeek V3.2
2026.06
53.3
20,632.6
Qwen 3 235B
Parameters=235B
2026.06
44.8
19,033
Feedback
Search any
task
Search any
task