Puzzle Solving on PUZZLE (Human easy)
69.5AccuracyGPT-5
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-52026.06 | 69.5 | 10,179.6 | |
| Kimi K22026.06 | 56.2 | 29,713.7 | |
| DeepSeek V3.22026.06 | 53.3 | 20,632.6 | |
| Qwen 3 235BParameters=235B2026.06 | 44.8 | 19,033 |
| Method | Links | ||
|---|---|---|---|
| GPT-52026.06 | 69.5 | 10,179.6 | |
| Kimi K22026.06 | 56.2 | 29,713.7 | |
| DeepSeek V3.22026.06 | 53.3 | 20,632.6 | |
| Qwen 3 235BParameters=235B2026.06 | 44.8 | 19,033 |