List your company
ResearchFormerly WizWand

Step-level Reasoning Verification on GSM8k

93PR-AUCQwen2.5-Math-PRM-7B
0.23224.31648.472.484Nov 9, 2025
Updated 6mo ago

Evaluation results

MethodLinks
Qwen2.5-Math-PRM-7B# Sample=860K, Verifying Model=Qwen3-32B, Thinking Mode=native2025.1193
Universal-PRM-Qwen2.5-Math-7B# Sample=690K, Verifying Model=Qwen3-32B, Thinking Mode=native2025.1190.3
Qwen2.5-Math-7B-PRM800K# Sample=263K, Verifying Model=Qwen3-32B, Thinking Mode=native2025.1189.4
Skywork-PRM-1.5B# Sample=Unk, Verifying Model=Qwen3-32B, Thinking Mode=native2025.1188.6
ReProbe, Hidden States, GPT-OSS-anno# Sample=10.8K, Verifying Model=Qwen3-32B, Thinking Mode=native2025.1153.5
H4-Qwen2.5-PRM-1.5B-0.2# Sample=369K, Verifying Model=Qwen3-32B, Thinking Mode=native2025.1146.4
Qwen2.5-Math-7B-PRM800k# Sample=265K2025.1140.6
Qwen2.5-Math-7B# Sample=860K2025.1137.7
ReProbe, Attn+Logit, Qwen3-8B-anno# Sample=32K2025.1134
RLHFlow-PRM-Deepseek-8B# Sample=253K2025.1126.3
Universal-PRM-Qwen2.5-Math-7B# Sample=690K2025.1121.3
RLHFlow-PRM-Mistral-8B# Sample=273K2025.1119.5
Math-Shepherd-PRM-7B# Sample=440K2025.1118.8
Skywork-PRM-1.5B# Sample=Unk2025.1118.1
MaxProb# Sample=-2025.118.4
MaxEntropy# Sample=-2025.117.9
Perplexity# Sample=-2025.116.6
H4-Qwen2.5-PRM-1.5B-0.2# Sample=369K2025.116.1
Random# Sample=-2025.113.8