ResearchBenchmarksMathematical Reasoning on O-bench (AVG@8)Follow45.54AVG@8IAC42.534443.314744.09544.8753Jan 22, 2026Evaluation ResultsMethodMethodLinksAVG@8IACModel Backbone=DeepSee...Model Backbone=DeepSeek-R1-Distill-Qwen-1.5B, Training Dataset=DAPO-Math-17K2026.0145.54DAPOModel Backbone=DeepSee...Model Backbone=DeepSeek-R1-Distill-Qwen-1.5B, Training Dataset=DAPO-Math-17K2026.0143.64GRPOModel Backbone=DeepSee...Model Backbone=DeepSeek-R1-Distill-Qwen-1.5B, Training Dataset=DAPO-Math-17K2026.0142.65