List your company

Mathematical Reasoning on MATH 500 (Pass@4, Pass@16, Pass@64)

66.8Pass@4Q-RM
8.14423.37238.653.828Apr 14, 2026

Evaluation results

MethodLinks
Q-RMBase Model=Qwen3-8B Base2026.0466.86968.8
IPVRMBase Model=Qwen3-8B Base2026.0466.867.669.2
DPO-RMBase Model=Qwen3-8B Base2026.0466.66869.8
Implicit PRMBase Model=Qwen3-8B Base2026.0466.46870.2
EndoRMBase Model=Qwen3-8B Base2026.0460.66157.6
IPVRMBase Model=Qwen3-0.6B Base2026.04424647.6
DPO-RMBase Model=Qwen3-0.6B Base2026.0441.642.642.2
Q-RMBase Model=Qwen3-0.6B Base2026.0441.444.444.6
Implicit PRMBase Model=Qwen3-0.6B Base2026.0439.639.638.6
EndoRMBase Model=Qwen3-0.6B Base2026.043233.838.8
IPVRMBase Model=Llama3.2-3B2026.0430.634.635.8
Implicit PRMBase Model=Llama3.2-3B2026.0429.832.835.4
DPO-RMBase Model=Llama3.2-3B2026.0429.231.634.6
Q-RMBase Model=Llama3.2-3B2026.042931.433.8
EndoRMBase Model=Llama3.2-3B2026.0425.426.827
IPVRMBase Model=Llama3.2-1B2026.0413.914.214.8
Q-RMBase Model=Llama3.2-1B2026.0413.614.613.4
Implicit PRMBase Model=Llama3.2-1B2026.04131414.2
DPO-RMBase Model=Llama3.2-1B2026.0411.212.612.6
EndoRMBase Model=Llama3.2-1B2026.0410.410.610