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

Process-outcome alignment on PRIME Physics

92.41Outcome Score

GPT-5.2-thinking

Updated 4d ago

Evaluation Results

Method	Links
GPT-5.2-thinking 2026.02		92.41	89.96
GPT-5.2-chat 2026.02		92.19	88.17
Gemini-2.5-Pro-thinking 2026.02		91.52	88.17
Claude-Sonnet-4.5-thinking 2026.02		91.07	89.96
Claude-Opus-4-thinking 2026.02		91.07	87.28
GPT-5-thinking 2026.02		90.85	89.96
Kimi-K2-thinking 2026.02		90.4	87.05
Claude-Opus-4.5 2026.02		90.18	81.47
DeepSeek-V3.2-thinking 2026.02		89.73	87.72
Claude-Sonnet-4-thinking 2026.02		89.73	85.94
Claude-Opus-4.5-thinking 2026.02		89.73	86.61
Qwen3-4B-thinking-2507 2026.02		89.51	82.81
GLM-4.6 2026.02		89.51	85.27
Grok-4-thinking 2026.02		89.51	87.72
GPT-OSS-20B 2026.02		88.62	83.26
CompassVerifier-32B 2026.02		88.62	77.46
GPT-OSS-120B 2026.02		88.39	82.37
Claude-Sonnet-4 2026.02		88.39	80.58
Claude-Opus-4 2026.02		88.39	81.47
Gemini-3.0-pro-thinking 2026.02		88.39	85.49
Qwen3-8B 2026.02		88.17	80.13
Qwen3-4B 2026.02		87.72	79.46
Seed-1.6-thinking 2026.02		87.72	79.91
DeepSeek-V3.2 2026.02		87.5	79.02
Qwen3-30B-A3B 2026.02		87.28	83.26
SCI-Verifier-4B 2026.02		87.05	77.68
Claude-Sonnet-4.5 2026.02		86.83	79.69
GPT-5-chat 2026.02		86.61	77.23
Tencent-Qwen2.5-7B-RLVR 2026.02		82.14	72.77
xVerify-9B-C 2026.02		81.92	73.44
Phi-4 2026.02		81.7	70.09
General-Verifier 2026.02		78.57	68.53
Qwen3-4B-Instruct-2507 2026.02		78.35	68.97
Math-Verify 2026.02		63.17	65.85