Poker Gameplay Performance on Limit Texas Hold'em
60.5NFSP PerformanceToolPoker
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| ToolPokerBackbone=Qwen2.5-7B2026.01 | 60.5 | 63 | 61.5 | -5 | 45 | |
| Qwen2.5-72BType=Vanilla2026.01 | 6 | -53.5 | -73.5 | -57.5 | -44.6 | |
| Qwen2.5-7BType=Vanilla2026.01 | -53.5 | -188 | -144 | -101 | -121.6 | |
| BC-RIRLBackbone=Qwen2.5-7B2026.01 | -77.5 | -82.5 | -80.5 | -70.2 | -77.6 | |
| o4-miniType=Vanilla2026.01 | -105 | 111 | -58 | -117 | -42.2 |