YZ Index
评测数据
当前展示:Run #432 | 2026-11-18 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-6 Astra gpt | 100.00 | 80.30 | 75.00 | 85.00 | 80.00 pass | 91.14 | 41.3 | 42.9 | 100.0 | |
| GPT-6.1 Sol gpt | 100.00 | 78.10 | 75.00 | 90.00 | 80.00 pass | 90.15 | 97.5 | 45.3 | 100.0 | |
| GPT-6 Luna gpt | 75.00 | 94.80 | 63.90 | 83.80 | 80.00 pass | 83.91 | 100.0 | 37.7 | 100.0 | |
| Grok 4 grok | 79.40 | 89.40 | 61.10 | 76.30 | 80.00 pass | 83.90 | 89.3 | 37.7 | 100.0 | |
| Gemini 2.5 Pro gemini | 75.00 | 91.70 | 50.00 | 80.00 | 90.00 pass | 82.52 | 96.0 | 37.4 | 95.0 | |
| DeepSeek V4 Pro deepseek | 75.00 | 89.40 | 50.00 | 75.00 | 90.00 pass | 81.48 | 98.9 | 49.5 | 85.0 | |
| GPT-5.5 gpt | 75.00 | 83.30 | 75.00 | 90.00 | 65.00 pass | 78.74 | 69.3 | 25.6 | 100.0 | |
| GPT-6 Sol gpt | 75.00 | 78.10 | 75.00 | 86.30 | 90.00 pass | 76.40 | 94.5 | 35.2 | 100.0 | |
| Claude Opus 4.7 claude | 70.00 | 83.30 | 50.00 | 40.00 | 90.00 pass | 75.99 | 21.6 | 19.9 | 100.0 | |
| GPT-o3 gpt | 50.00 | 100.00 | 75.00 | 90.00 | 80.00 pass | 72.50 | 42.5 | 36.1 | 95.0 | |
| 豆包 Pro doubao | 75.00 | 67.90 | 75.00 | 95.00 | 90.00 pass | 71.81 | 100.0 | 34.0 | 100.0 | |
| GLM-4.6 zhipu | 50.00 | 97.80 | 8.30 | 15.00 | 65.00 pass | 71.51 | 94.7 | 25.6 | 65.0 | |
| Claude Sonnet 4.6 claude | 55.20 | 75.50 | 50.00 | 51.30 | 80.00 pass | 64.34 | 85.4 | 25.8 | 100.0 | |
| Gemini 3.1 Pro gemini | 50.00 | 67.90 | 50.00 | 90.00 | 90.00 pass | 58.06 | 79.7 | 20.0 | 100.0 | |
| 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。