YZ Index
评测数据
当前展示:Run #280 | 2026-08-16 | 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 | 75.00 | 91.70 | 100.00 pass | 86.25 | 27.7 | 39.4 | 100.0 | |
| GPT-5.5 gpt | 75.00 | 100.00 | 75.00 | 91.70 | 90.00 pass | 86.25 | 76.7 | 38.4 | 100.0 | |
| GPT-o3 gpt | 75.00 | 100.00 | 52.30 | 91.70 | 80.00 pass | 86.25 | 45.2 | 30.1 | 100.0 | |
| Grok 4 grok | 75.00 | 100.00 | 25.00 | 80.80 | 90.00 pass | 86.25 | 86.3 | 19.8 | 100.0 | |
| Claude Sonnet 4.6 claude | 75.00 | 75.00 | 60.50 | 70.00 | 90.00 pass | 75.00 | 83.6 | 20.1 | 100.0 | |
| Qwen3 Max qwen | 75.00 | 75.00 | 25.00 | 55.60 | 90.00 pass | 75.00 | 98.7 | 11.8 | 100.0 | |
| DeepSeek V4 Pro deepseek | 66.70 | 75.00 | 75.00 | 66.70 | 90.00 pass | 70.44 | 97.0 | 48.4 | 85.0 | |
| 豆包 Pro doubao | 66.70 | 75.00 | 75.00 | 86.70 | 90.00 pass | 70.44 | 100.0 | 37.4 | 95.0 | |
| Gemini 3.1 Pro gemini | 66.70 | 75.00 | 75.00 | 66.70 | 90.00 pass | 70.44 | 83.3 | 21.3 | 100.0 | |
| Gemini 2.5 Pro gemini | 37.50 | 75.00 | 75.00 | 41.70 | 90.00 pass | 54.38 | 89.5 | 23.8 | 85.0 | |
| GLM-4.6 zhipu | 37.50 | 62.50 | 50.00 | 25.00 | 80.00 pass | 48.75 | 87.3 | 22.0 | 75.0 |
API 访问:如需程序化访问评测数据,请使用我们的 API。