List your company

Reasoning on Aggregate (AIME’24, AIME’25, MATH 500, AMC’23, GPQA Diamond) (test)

14.64AAIPPPO
1.24484.72248.211.6776Dec 17, 2025

Evaluation results

MethodLinks
PPPOBackbone=Qwen3-8B2025.1214.640.248358.95
PPPOBackbone=Qwen3-4B2025.1212.360.261747.24
PPPOBackbone=Qwen3-1.7B2025.1210.520.291936.05
DAPO-FTBackbone=Qwen3-8B2025.129.470.247.36
DAPOBackbone=Qwen3-8B2025.129.1919.19
DAPOBackbone=Qwen3-4B2025.127.3917.39
DAPO-FTBackbone=Qwen3-4B2025.127.390.237.02
DAPO-FTBackbone=Qwen3-1.7B2025.125.620.228.08
DAPOBackbone=Qwen3-1.7B2025.125.4615.46
GRPOBackbone=Qwen3-8B2025.124.7414.74
GRPOBackbone=Qwen3-4B2025.123.3713.37
INTUITORBackbone=Qwen3-8B2025.123.3413.34
GRPOBackbone=Qwen3-1.7B2025.122.8712.87
INTUITORBackbone=Qwen3-4B2025.122.4212.42
INTUITORBackbone=Qwen3-1.7B2025.121.7611.76