YZ Index
评测数据
当前展示:Run #358 | 2026-10-03 | 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 | 100.00 | 97.00 | 100.00 | 55.60 | 100.00 pass | 98.65 | 30.4 | 52.2 | 100.0 | |
| GPT-o3 gpt | 100.00 | 83.80 | 100.00 | 91.70 | 90.00 pass | 92.71 | 50.5 | 69.4 | 100.0 | |
| Gemini 2.5 Pro gemini | 100.00 | 72.00 | 100.00 | 91.70 | 100.00 pass | 87.40 | 98.3 | 54.8 | 100.0 | |
| Gemini 3.1 Pro gemini | 100.00 | 72.00 | 75.00 | 91.70 | 100.00 pass | 87.40 | 90.3 | 39.6 | 100.0 | |
| GPT-6 Sol gpt | 75.00 | 97.80 | 100.00 | 87.50 | 100.00 pass | 85.26 | 97.8 | 52.4 | 100.0 | |
| Grok 4 grok | 75.00 | 97.00 | 47.20 | 91.70 | 100.00 pass | 84.90 | 88.5 | 33.6 | 100.0 | |
| DeepSeek V4 Pro deepseek | 75.00 | 94.80 | 75.00 | 91.70 | 90.00 pass | 83.91 | 98.8 | 75.7 | 90.0 | |
| Claude Sonnet 4.6 claude | 73.80 | 86.80 | 77.20 | 70.80 | 100.00 pass | 79.65 | 87.4 | 35.4 | 100.0 | |
| GPT-6.1 Sol gpt | 58.30 | 100.00 | 100.00 | 87.50 | 100.00 pass | 77.07 | 97.1 | 46.0 | 100.0 | |
| GPT-6 Astra gpt | 58.30 | 94.80 | 100.00 | 87.50 | 100.00 pass | 74.73 | 39.3 | 46.0 | 100.0 | |
| 豆包 Pro doubao | 83.30 | 58.80 | 100.00 | 91.70 | 90.00 pass | 72.28 | 100.0 | 44.4 | 100.0 | |
| GPT-5.5 gpt | 58.30 | 81.50 | 100.00 | 91.70 | 100.00 pass | 68.74 | 73.9 | 45.1 | 100.0 | |
| GPT-6 Luna gpt | 58.30 | 81.50 | 80.90 | 87.50 | 100.00 pass | 68.74 | 100.0 | 41.9 | 100.0 | |
| Qwen3 Max qwen | 75.00 | 56.00 | 16.80 | 35.60 | 90.00 pass | 66.45 | 97.7 | 20.8 | 100.0 | |
| GLM-4.6 zhipu | 50.00 | 25.00 | 50.00 | 37.50 | 50.00 warn | 38.75 | 83.6 | 68.6 | 45.0 |
API 访问:如需程序化访问评测数据,请使用我们的 API。