Puzzle Solving on PUZZLE Human hard
5.7AccuracyGPT-5
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-52026.06 | 5.7 | 19,861.9 | |
| Kimi K22026.06 | 1 | 61,307.1 | |
| Qwen 3 235BParameters=235B2026.06 | 0 | 23,608.6 | |
| DeepSeek V3.22026.06 | 0 | 36,787.4 |
| Method | Links | ||
|---|---|---|---|
| GPT-52026.06 | 5.7 | 19,861.9 | |
| Kimi K22026.06 | 1 | 61,307.1 | |
| Qwen 3 235BParameters=235B2026.06 | 0 | 23,608.6 | |
| DeepSeek V3.22026.06 | 0 | 36,787.4 |