Mathematical Reasoning on AIME 25 (Speedup, Avg Tau)
8.78SpeedupJetFlow
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| JetFlowBudget=256, Temperature=02026.06 | 8.78 | 9.82 | |
| DFlash-TBudget=256, Temperature=02026.06 | 8.33 | 9.24 | |
| JetFlowBudget=128, Temperature=02026.06 | 8.26 | 9.1 | |
| DFlash-TBudget=128, Temperature=02026.06 | 7.93 | 8.66 | |
| JetFlowBudget=64, Temperature=02026.06 | 6.47 | 7 | |
| DFlash-TBudget=64, Temperature=02026.06 | 6.4 | 6.96 | |
| JetFlowBudget=256, Temperature=12026.06 | 5.94 | 7.06 | |
| JetFlowBudget=128, Temperature=12026.06 | 5.76 | 6.49 | |
| DFlash-TBudget=128, Temperature=12026.06 | 5.4 | 5.98 | |
| DFlash-TBudget=256, Temperature=12026.06 | 5.26 | 6.2 | |
| JetFlowBudget=64, Temperature=12026.06 | 4.95 | 5.48 | |
| DFlash-TBudget=64, Temperature=12026.06 | 4.94 | 5.46 | |
| EAGLE-3Budget=64, Temperature=02026.06 | 2.35 | 4.04 | |
| EAGLE-3Budget=64, Temperature=12026.06 | 2.11 | 3.73 |