YZ Index
评测数据
当前展示：Run #356 | 2026-10-02 | 128 题库 | 公式 v7 | 判分 v6.4
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 |
|---|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.7 claude | 75.00 | 100.00 | 100.00 | 46.70 | 90.00 pass | 86.25 | 26.8 | 37.3 | 100.0 | |
| GPT-5.5 gpt | 75.00 | 100.00 | 100.00 | 81.70 | 65.00 pass | 86.25 | 76.2 | 37.7 | 100.0 | |
| GPT-6 Astra gpt | 75.00 | 100.00 | 100.00 | 72.50 | 70.00 pass | 86.25 | 39.3 | 44.4 | 100.0 | |
| GPT-6.1 Sol gpt | 75.00 | 100.00 | 100.00 | 72.50 | 70.00 pass | 86.25 | 96.9 | 44.4 | 100.0 | |
| Gemini 2.5 Pro gemini | 71.90 | 100.00 | 100.00 | 81.70 | 90.00 pass | 84.55 | 97.8 | 51.2 | 100.0 | |
| GPT-6 Sol gpt | 75.00 | 95.00 | 100.00 | 77.50 | 80.00 pass | 84.00 | 96.9 | 48.0 | 100.0 | |
| Grok 4 grok | 70.30 | 100.00 | 75.00 | 81.70 | 70.00 pass | 83.67 | 86.7 | 39.1 | 100.0 | |
| Claude Sonnet 4.6 claude | 75.00 | 90.00 | 100.00 | 95.00 | 80.00 pass | 81.75 | 88.2 | 50.4 | 100.0 | |
| 豆包 Pro doubao | 87.00 | 75.00 | 100.00 | 86.70 | 80.00 pass | 81.60 | 100.0 | 50.8 | 100.0 | |
| GPT-6 Luna gpt | 75.00 | 85.00 | 60.90 | 81.70 | 55.00 warn | 79.50 | 100.0 | 38.4 | 100.0 | |
| GPT-o3 gpt | 70.30 | 75.00 | 100.00 | 81.70 | 65.00 pass | 72.42 | 41.2 | 28.8 | 100.0 | |
| Qwen3 Max qwen | 71.90 | 70.00 | 36.30 | 51.70 | 75.00 pass | 71.05 | 98.3 | 19.7 | 100.0 | |
| DeepSeek V4 Pro deepseek | 41.70 | 100.00 | 75.00 | 81.70 | 80.00 pass | 67.94 | 96.6 | 65.8 | 85.0 | |
| GLM-4.6 zhipu | 41.70 | 100.00 | 75.00 | 82.50 | 15.00 fail | 67.94 | 95.0 | 44.2 | 75.0 | |
| Gemini 3.1 Pro gemini | 43.40 | 75.00 | 75.00 | 76.70 | 90.00 pass | 57.62 | 77.7 | 22.0 | 100.0 |
API 访问：如需程序化访问评测数据，请使用我们的 API。