List your company
ResearchFormerly WizWand

Mathematical Reasoning on AIME'24 (accuracy avg@32)

72.19Accuracy (avg@32)PPPO
23.975636.492849.0161.5272Dec 17, 2025
Updated 8mo ago

Evaluation results

MethodLinks
PPPOBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1272.19
DAPO-FTBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1263.75
PPPOBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1263.54
DAPOBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1263.13
GRPOBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1258.75
DAPOBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1256.46
DAPO-FTBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1256.25
INTUITORBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1255.42
Qwen3-8BBackbone=Qwen3-8B, Evaluation Protocol=zero-shot, avg@322025.1252.29
GRPOBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1252.08
INTUITORBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1251.04
Qwen3-4BBackbone=Qwen3-4B, Evaluation Protocol=zero-shot, avg@322025.1248.75
PPPOBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1238.65
DAPO-FTBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1231.25
DAPOBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1230.83
GRPOBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1227.5
INTUITORBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1227.29
Qwen3-1.7BBackbone=Qwen3-1.7B, Evaluation Protocol=zero-shot, avg@322025.1225.83