List your company
ResearchFormerly WizWand

Mathematical Reasoning on AIME'25 (Accuracy Avg@32)

59.69Accuracy (Avg@32)PPPO
16.030827.365438.750.0346Dec 17, 2025
Updated 8mo ago

Evaluation results

MethodLinks
PPPOBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1259.69
PPPOBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1253.44
DAPO-FTBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1249.38
DAPOBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1248.75
GRPOBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1242.29
DAPOBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1242.08
DAPO-FTBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1242.08
INTUITORBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1240.83
Qwen3-8BBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1238.75
GRPOBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1237.71
Qwen3-4BBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1235.42
INTUITORBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1235.42
PPPOBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1228.96
DAPO-FTBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1223.96
DAPOBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1223.33
GRPOBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1220
Qwen3-1.7BBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1218.33
INTUITORBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1217.71