YZ Index
评测数据
当前展示:Run #366 | 2026-10-08 | 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 Sonnet 4.6 claude | 98.70 | 73.60 | 100.00 | 83.90 | 80.00 pass | 87.41 | 89.5 | 59.6 | 100.0 | |
| GPT-6 Luna gpt | 75.00 | 93.20 | 100.00 | 90.00 | 80.00 pass | 83.19 | 100.0 | 50.0 | 100.0 | |
| Claude Opus 4.7 claude | 75.00 | 84.10 | 100.00 | 53.90 | 100.00 pass | 79.10 | 26.4 | 39.4 | 100.0 | |
| GPT-o3 gpt | 75.00 | 76.40 | 100.00 | 90.00 | 90.00 pass | 75.63 | 45.4 | 46.9 | 100.0 | |
| GPT-5.5 gpt | 75.00 | 73.60 | 100.00 | 90.00 | 80.00 pass | 74.37 | 74.2 | 44.7 | 100.0 | |
| Grok 4 grok | 70.30 | 65.90 | 80.60 | 90.00 | 80.00 pass | 68.32 | 84.0 | 34.5 | 100.0 | |
| GPT-6 Sol gpt | 75.00 | 50.00 | 100.00 | 90.00 | 80.00 pass | 63.75 | 94.6 | 27.9 | 100.0 | |
| Qwen3 Max qwen | 71.00 | 51.80 | 36.30 | 53.90 | 90.00 pass | 62.36 | 98.0 | 28.7 | 100.0 | |
| GPT-6 Astra gpt | 50.00 | 75.00 | 100.00 | 85.00 | 80.00 pass | 61.25 | 34.1 | 26.4 | 100.0 | |
| GPT-6.1 Sol gpt | 50.00 | 75.00 | 100.00 | 90.00 | 80.00 pass | 61.25 | 94.6 | 27.9 | 100.0 | |
| 豆包 Pro doubao | 50.00 | 73.60 | 100.00 | 95.00 | 80.00 pass | 60.62 | 100.0 | 44.8 | 95.0 | |
| Gemini 2.5 Pro gemini | 48.70 | 73.60 | 75.00 | 90.00 | 90.00 pass | 59.91 | 95.1 | 43.6 | 90.0 | |
| Gemini 3.1 Pro gemini | 50.00 | 69.60 | 100.00 | 85.00 | 80.00 pass | 58.82 | 83.0 | 31.4 | 100.0 | |
| DeepSeek V4 Pro deepseek | 25.00 | 93.20 | 100.00 | 95.00 | 80.00 pass | 55.69 | 97.0 | 49.5 | 90.0 | |
| GLM-4.6 zhipu | 25.00 | 34.10 | 25.00 | 95.00 | 90.00 pass | 29.10 | 85.9 | 38.7 | 65.0 |
API 访问:如需程序化访问评测数据,请使用我们的 API。