YZ Index
评测数据
当前展示：Run #418 | 2026-11-09 | 128 题库 | 公式 v7 | 判分 v6.5
数据公开说明：为防止题库污染和过拟合，题目原文和预期答案不公开。本页展示模型回答、得分、判分方式等透明数据。完整方法论请参阅方法论页面。
| 模型 | 代码执行 | 材料约束 | 工程判断 | 任务表达 | 诚信 | 主榜分 | 性价比 | 稳定性 | 可用性 | 逐题 |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-6 Sol gpt | 100.00 | 83.30 | 57.50 | 100.00 | 100.00 pass | 92.49 | 97.8 | 42.5 | 100.0 | |
| GPT-o3 gpt | 100.00 | 83.30 | 57.50 | 75.00 | 90.00 pass | 92.49 | 47.4 | 43.7 | 95.0 | |
| Grok 4 grok | 93.80 | 88.30 | 25.00 | 71.30 | 90.00 pass | 91.33 | 87.6 | 21.8 | 100.0 | |
| Claude Sonnet 4.6 claude | 100.00 | 80.00 | 50.00 | 70.00 | 100.00 pass | 91.00 | 88.7 | 33.8 | 100.0 | |
| GPT-6 Luna gpt | 100.00 | 78.30 | 75.00 | 100.00 | 100.00 pass | 90.24 | 100.0 | 48.8 | 100.0 | |
| GPT-6 Astra gpt | 100.00 | 73.30 | 75.00 | 100.00 | 100.00 pass | 87.99 | 41.5 | 47.3 | 100.0 | |
| GPT-6.1 Sol gpt | 100.00 | 73.30 | 75.00 | 100.00 | 100.00 pass | 87.99 | 97.6 | 47.3 | 100.0 | |
| 豆包 Pro doubao | 100.00 | 63.30 | 69.50 | 96.30 | 80.00 pass | 83.49 | 100.0 | 39.8 | 100.0 | |
| Gemini 2.5 Pro gemini | 75.00 | 91.70 | 50.00 | 81.30 | 100.00 pass | 82.52 | 97.4 | 47.2 | 90.0 | |
| Claude Opus 4.7 claude | 75.00 | 83.30 | 55.60 | 50.00 | 90.00 pass | 78.74 | 24.3 | 22.8 | 100.0 | |
| Gemini 3.1 Pro gemini | 75.00 | 83.30 | 75.00 | 75.00 | 100.00 pass | 78.74 | 88.1 | 25.5 | 100.0 | |
| GPT-5.5 gpt | 75.00 | 83.30 | 75.00 | 100.00 | 100.00 pass | 78.74 | 76.2 | 35.4 | 100.0 | |
| DeepSeek V4 Pro deepseek | 50.00 | 90.00 | 75.00 | 75.00 | 90.00 pass | 68.00 | 97.6 | 73.5 | 80.0 | |
| GLM-4.6 zhipu | 25.00 | 50.00 | 25.00 | 25.00 | 65.00 pass | 36.25 | 81.2 | 35.3 | 45.0 | |
| Qwen3 Max qwen | - | - | - | - | - | - | 0.0 | 75.0 | 0.0 |
API 访问：如需程序化访问评测数据，请使用我们的 API。