List your company
ResearchFormerly WizWand

Mathematical Reasoning on Math500 decontaminated (Accuracy)

77.8AccuracyDAPO-Math-17k
47.22455.16263.171.038May 26, 2026
Updated 5mo ago

Evaluation results

MethodLinks
DAPO-Math-17kBackbone=Qwen3-8B, RLVR Training Data=DAPO-Math-17k, Sampling Strategy=Mean@42026.0577.8
DAPO++Backbone=Qwen3-8B, Sampling Strategy=Mean@42026.0577.7
DeepScaleRBackbone=Qwen3-8B, RLVR Training Data=DeepScaleR, Sampling Strategy=Mean@42026.0575.9
DeepMath-103KBackbone=Qwen3-8B, RLVR Training Data=DeepMath-103K, Sampling Strategy=Mean@42026.0573.4
Skywork-OR1-RL-DataBackbone=Qwen3-8B, RLVR Training Data=Skywork-OR1-RL-Data, Sampling Strategy=Mean@42026.0573.2
OpenR1-Math-220kBackbone=Qwen3-8B, RLVR Training Data=OpenR1-Math-220k, Sampling Strategy=Mean@42026.0573.1
Qwen3-8B-BaseBackbone=Qwen3-8B, Sampling Strategy=Mean@42026.0561.6
DAPO++Backbone=Qwen3-1.7B, Sampling Strategy=Mean@42026.0560
OpenR1-Math-220kBackbone=Qwen3-1.7B, RLVR Training Data=OpenR1-Math-220k, Sampling Strategy=Mean@42026.0558.9
Skywork-OR1-RL-DataBackbone=Qwen3-1.7B, RLVR Training Data=Skywork-OR1-RL-Data, Sampling Strategy=Mean@42026.0558.8
DAPO-Math-17kBackbone=Qwen3-1.7B, RLVR Training Data=DAPO-Math-17k, Sampling Strategy=Mean@42026.0558.5
DeepScaleRBackbone=Qwen3-1.7B, RLVR Training Data=DeepScaleR, Sampling Strategy=Mean@42026.0558.1
DeepMath-103KBackbone=Qwen3-1.7B, RLVR Training Data=DeepMath-103K, Sampling Strategy=Mean@42026.0557.9
Qwen3-1.7B-BaseBackbone=Qwen3-1.7B, Sampling Strategy=Mean@42026.0548.4