Reasoning Evaluation on Leduc Hold’em
2Hit Rate (HR)Professional
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Professional2026.01 | 2 | 2 | 2 | 2 | |
| o4-mini2026.01 | 1.8 | 1.56 | 1.85 | 1.73 | |
| Qwen2.5-72BModel Size=72B2026.01 | 1.03 | 1.23 | 1.78 | 1.34 | |
| GPT-4.1-mini2026.01 | 0.98 | 1.73 | 1.87 | 1.52 | |
| Qwen2.5-7BModel Size=7B2026.01 | 0.95 | 0.86 | 1.68 | 1.16 | |
| Qwen2.5-3BModel Size=3B2026.01 | 0.53 | 0.18 | 1.53 | 0.74 |