Process-outcome alignment on PRIME Average
91.81Outcome ScoreClaude-Sonnet-4.5-thinking
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Claude-Sonnet-4.5-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 91.81 | 85.99 | 84.45 | |
| GPT-5.2-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 91.57 | 87.97 | 86.36 | |
| Gemini-2.5-Pro-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 91.52 | 88.56 | 86.19 | |
| DeepSeek-V3.2-thinkingCategory=Open-source Models, Thinking=true2026.02 | 91.41 | 87.02 | 85.65 | |
| Kimi-K2-thinkingCategory=Open-source Models, Thinking=true2026.02 | 91.41 | 86.15 | 84.6 | |
| GPT-5.2-chatCategory=Closed-source Models, Thinking=false2026.02 | 91.29 | 85.6 | 84.32 | |
| Grok-4-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 90.98 | 87.69 | 85.75 | |
| Claude-Opus-4-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 90.94 | 79.5 | 83.27 | |
| Qwen3-4B-thinking-2507Category=Open-source Models, Thinking=true2026.02 | 90.15 | 81.36 | 79.88 | |
| GPT-5-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 90.15 | 88.09 | 86.71 | |
| GPT-OSS-120BCategory=Open-source Models, Thinking=false2026.02 | 89.99 | 81.64 | 80.71 | |
| Claude-Sonnet-4-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 89.79 | 83.89 | 82.75 | |
| Claude-Opus-4.5Category=Closed-source Models, Thinking=false2026.02 | 89.55 | 78.91 | 78.86 | |
| GLM-4.6Category=Open-source Models, Thinking=false2026.02 | 89.32 | 83.5 | 82.26 | |
| Qwen3-30B-A3BCategory=Open-source Models, Thinking=false2026.02 | 89.16 | 82.71 | 80.34 | |
| Gemini-3.0-pro-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 89.08 | 84.33 | 83.35 | |
| GPT-OSS-20BCategory=Open-source Models, Thinking=false2026.02 | 89.04 | 79.74 | 78.75 | |
| Claude-Opus-4Category=Closed-source Models, Thinking=false2026.02 | 89 | 79.5 | 78.38 | |
| Seed-1.6-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 88.8 | 78.51 | 78.5 | |
| CompassVerifier-32BCategory=Verifier Model, Thinking=false2026.02 | 88.68 | 73.29 | 73.17 | |
| Claude-Sonnet-4Category=Closed-source Models, Thinking=false2026.02 | 88.25 | 77.8 | 77.06 | |
| Claude-Opus-4.5-thinkingCategory=Closed-source Models, Thinking=true2026.02 | 88.21 | 82.31 | 82.13 | |
| SCI-Verifier-4BCategory=Verifier Model, Thinking=false2026.02 | 88.17 | 74.28 | 73.77 | |
| Claude-Sonnet-4.5Category=Closed-source Models, Thinking=false2026.02 | 88.05 | 76.97 | 76.79 | |
| DeepSeek-V3.2Category=Open-source Models, Thinking=false2026.02 | 87.81 | 77.13 | 76.33 | |
| Qwen3-8BCategory=Open-source Models, Thinking=false2026.02 | 87.77 | 77.32 | 76.62 | |
| Qwen3-4BCategory=Open-source Models, Thinking=false2026.02 | 86.43 | 75.23 | 74.26 | |
| GPT-5-chatCategory=Closed-source Models, Thinking=false2026.02 | 85.4 | 72.42 | 74.02 | |
| xVerify-9B-CCategory=Verifier Model, Thinking=false2026.02 | 84.92 | 71.98 | 70.91 | |
| Tencent-Qwen2.5-7B-RLVRCategory=Verifier Model, Thinking=false2026.02 | 84.84 | 70.87 | 70.17 | |
| Phi-4Category=Open-source Models, Thinking=false2026.02 | 80.29 | 64.46 | 69.08 | |
| Qwen3-4B-Instruct-2507Category=Open-source Models, Thinking=false2026.02 | 79.74 | 73.13 | 68.74 | |
| General-VerifierCategory=Verifier Model, Thinking=false2026.02 | 77.13 | 63.04 | 68.12 | |
| Math-VerifyCategory=Rule-based Verifier, Thinking=false2026.02 | 67.71 | 65.93 | 57.88 |