YZ Index
评测数据
当前展示:Run #375 | 2026-10-13 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| Gemini 3.1 Pro gemini | 91.70 | 71.00 | 100.00 | 51.70 | 80.00 pass | 82.39 | 87.8 | 41.2 | 100.0 | |
| GPT-o3 gpt | 75.00 | 65.90 | 100.00 | 81.70 | 75.00 pass | 70.91 | 41.2 | 28.4 | 100.0 | |
| Grok 4 grok | 75.00 | 65.90 | 71.40 | 61.70 | 40.00 warn | 70.91 | 79.1 | 20.1 | 100.0 | |
| 豆包 Pro doubao | 75.00 | 61.90 | 100.00 | 71.70 | 80.00 pass | 69.11 | 100.0 | 50.1 | 100.0 | |
| DeepSeek V4 Pro deepseek | 50.00 | 81.70 | 100.00 | 77.50 | 65.00 pass | 64.27 | 96.2 | 38.3 | 95.0 | |
| GPT-6 Luna gpt | 75.00 | 48.00 | 75.00 | 81.70 | 80.00 pass | 62.85 | 100.0 | 34.0 | 100.0 | |
| GPT-6.1 Sol gpt | 58.30 | 64.30 | 100.00 | 67.20 | 70.00 pass | 61.00 | 91.8 | 33.8 | 100.0 | |
| Claude Sonnet 4.6 claude | 50.00 | 72.60 | 85.50 | 90.00 | 70.00 pass | 60.17 | 83.2 | 34.1 | 100.0 | |
| GPT-5.5 gpt | 58.30 | 61.90 | 100.00 | 77.50 | 70.00 pass | 59.92 | 64.1 | 40.2 | 100.0 | |
| GPT-6 Astra gpt | 58.30 | 61.90 | 100.00 | 67.20 | 70.00 pass | 59.92 | 30.2 | 39.9 | 100.0 | |
| GPT-6 Sol gpt | 75.00 | 41.30 | 100.00 | 72.20 | 90.00 pass | 59.84 | 92.7 | 35.8 | 100.0 | |
| Claude Opus 4.7 claude | 50.00 | 71.00 | 95.80 | 76.70 | 90.00 pass | 59.45 | 23.8 | 34.7 | 100.0 | |
| Gemini 2.5 Pro gemini | 50.00 | 65.90 | 75.10 | 36.70 | 70.00 pass | 57.16 | 89.5 | 29.9 | 90.0 | |
| GLM-4.6 zhipu | 50.00 | 50.00 | 75.00 | 36.70 | 75.00 pass | 50.00 | 90.2 | 30.5 | 70.0 | |
| 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。