YZ Index
评测数据
当前展示：Run #404 | 2026-10-31 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| Gemini 3.1 Pro gemini | 94.00 | 85.90 | 75.00 | 100.00 | 60.00 pass | 90.36 | 90.8 | 49.0 | 100.0 | |
| 豆包 Pro doubao | 81.50 | 97.00 | 75.00 | 100.00 | 70.00 pass | 88.48 | 100.0 | 48.9 | 100.0 | |
| GPT-6 Astra gpt | 94.00 | 77.20 | 75.00 | 94.70 | 70.00 pass | 86.44 | 40.6 | 48.3 | 100.0 | |
| Claude Sonnet 4.6 claude | 94.00 | 74.40 | 38.90 | 62.50 | 80.00 pass | 85.18 | 86.8 | 32.0 | 100.0 | |
| DeepSeek V4 Pro deepseek | 94.00 | 73.70 | 75.00 | 100.00 | 60.00 pass | 84.87 | 98.7 | 47.8 | 100.0 | |
| GPT-6 Luna gpt | 94.00 | 72.00 | 75.00 | 100.00 | 80.00 pass | 84.10 | 100.0 | 48.8 | 100.0 | |
| GPT-6.1 Sol gpt | 94.00 | 72.00 | 75.00 | 100.00 | 70.00 pass | 84.10 | 97.2 | 48.3 | 100.0 | |
| Grok 4 grok | 94.00 | 71.40 | 75.00 | 100.00 | 80.00 pass | 83.83 | 89.1 | 48.6 | 100.0 | |
| Claude Opus 4.7 claude | 69.00 | 100.00 | 62.50 | 50.00 | 90.00 pass | 82.95 | 24.0 | 25.9 | 100.0 | |
| Gemini 2.5 Pro gemini | 81.30 | 74.20 | 75.00 | 65.00 | 100.00 pass | 78.11 | 96.0 | 43.0 | 95.0 | |
| GPT-o3 gpt | 69.00 | 88.30 | 75.00 | 100.00 | 90.00 pass | 77.69 | 42.5 | 38.0 | 100.0 | |
| GPT-6 Sol gpt | 66.80 | 83.10 | 75.00 | 91.00 | 90.00 pass | 74.14 | 94.9 | 39.4 | 100.0 | |
| GPT-5.5 gpt | 69.00 | 72.00 | 75.00 | 100.00 | 80.00 pass | 70.35 | 68.0 | 37.7 | 100.0 | |
| GLM-4.6 zhipu | 72.00 | 13.30 | 25.00 | 50.00 | 90.00 pass | 45.59 | 93.1 | 66.8 | 55.0 | |
| 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。