YZ Index
评测数据
当前展示:Run #401 | 2026-10-29 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| Claude Sonnet 4.6 claude | 100.00 | 71.90 | 57.50 | 57.50 | 90.00 pass | 87.36 | 87.0 | 36.1 | 100.0 | |
| GPT-6 Astra gpt | 100.00 | 71.90 | 100.00 | 100.00 | 90.00 pass | 87.36 | 42.0 | 68.9 | 100.0 | |
| GPT-6 Sol gpt | 100.00 | 71.90 | 82.50 | 96.30 | 90.00 pass | 87.36 | 97.5 | 59.3 | 100.0 | |
| GPT-6.1 Sol gpt | 100.00 | 71.90 | 100.00 | 100.00 | 90.00 pass | 87.36 | 97.8 | 68.9 | 100.0 | |
| 豆包 Pro doubao | 100.00 | 66.90 | 100.00 | 100.00 | 80.00 pass | 85.11 | 100.0 | 63.1 | 100.0 | |
| GPT-o3 gpt | 75.00 | 82.60 | 71.90 | 75.00 | 65.00 pass | 78.42 | 42.7 | 31.7 | 95.0 | |
| Gemini 3.1 Pro gemini | 82.00 | 71.90 | 100.00 | 75.00 | 90.00 pass | 77.46 | 88.2 | 43.6 | 100.0 | |
| Claude Opus 4.7 claude | 75.00 | 71.90 | 72.20 | 57.50 | 80.00 pass | 73.61 | 23.6 | 38.0 | 100.0 | |
| DeepSeek V4 Pro deepseek | 75.00 | 71.90 | 100.00 | 75.00 | 90.00 pass | 73.61 | 98.2 | 67.7 | 90.0 | |
| Gemini 2.5 Pro gemini | 70.80 | 71.90 | 75.00 | 71.30 | 90.00 pass | 71.30 | 96.2 | 67.5 | 85.0 | |
| Grok 4 grok | 82.30 | 57.60 | 55.60 | 85.00 | 70.00 pass | 71.19 | 82.3 | 31.3 | 100.0 | |
| GPT-5.5 gpt | 75.00 | 65.20 | 100.00 | 100.00 | 90.00 pass | 70.59 | 73.9 | 46.7 | 100.0 | |
| GPT-6 Luna gpt | 75.00 | 57.60 | 100.00 | 75.00 | 90.00 pass | 67.17 | 100.0 | 28.2 | 100.0 | |
| GLM-4.6 zhipu | 50.00 | 71.90 | 100.00 | 25.00 | 80.00 pass | 59.86 | 95.6 | 63.0 | 75.0 | |
| 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。