Our new X account is live! Follow @wizwand_team for updates

Process Verification on ProcessBench With Standard Answers

78.9Precise Accuracy

OPV-32B

Updated 4d ago

Evaluation Results

Method	Links
OPV-32B 2025.12		78.9	78.9	79.1	79.1	83.3	82.7	87.2	86.2
OPV-Stage2 2025.12		71.6	71.6	75.9	75.9	76.2	79	81.1	82.6
R1-Distill-Qwen-32B 2025.12		70.5	70.5	76.3	76.3	74.9	79.1	78.7	81.7
OPV-Stage1 2025.12		68.8	68.8	75.7	75.7	72.6	78	76.1	80.2
Qwen3-32B 2025.12		68	68	73.4	73.4	72.4	76.2	77.8	79.9
DeepSeek-V3-0324 2025.12		67.9	67.9	71.7	71.7	72.9	75	78.6	79.1
DeepSeek-R1-0528 2025.12		67	67	71.2	71.2	72.3	74.6	78.3	79
Qwen3-Max-Preview 2025.12		66.4	66.4	71.2	71.2	71.7	74.6	78	79
QwQ-32B 2025.12		65.2	65.2	71.6	71.6	70.3	74.8	75.6	78.2
gpt-oss-120b (high) 2025.12		61	61	66.6	66.6	69.8	72.1	77.7	77.8
Qwen2.5-Math-PRM-72B 2025.12		55.4	55.4	66.5	66.5	58.4	70	66	72.3
gpt-oss-120b (high) 2025.12		0.833	0.833	0.788	0.788	0.892	0.851	0.957	0.935
Qwen3-Max-Preview 2025.12		0.832	0.832	0.789	0.789	0.894	0.855	0.954	0.932
DeepSeek-R1-0528 2025.12		0.822	0.822	0.777	0.777	0.887	0.846	0.954	0.931
OPV-32B 2025.12		0.801	0.801	0.762	0.762	0.876	0.838	0.962	0.944
Qwen2.5-Math-PRM-72B 2025.12		0.772	0.772	0.74	0.74	0.849	0.812	0.937	0.911
DeepSeek-V3-0324 2025.12		0.745	0.745	0.717	0.717	0.837	0.799	0.944	0.92
R1-Distill-Qwen-32B 2025.12		0.647	0.647	0.608	0.608	0.736	0.675	0.874	0.813