YZ Index
评测数据
当前展示:Run #378 | 2026-10-14 | 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 Opus 4.7 claude | 94.50 | 89.80 | 80.60 | 85.00 | 90.00 pass | 92.39 | 29.1 | 59.8 | 100.0 | |
| DeepSeek V4 Pro deepseek | 94.50 | 89.80 | 100.00 | 65.00 | 80.00 pass | 92.39 | 99.0 | 69.7 | 95.0 | |
| GPT-o3 gpt | 94.50 | 89.80 | 100.00 | 70.00 | 90.00 pass | 92.39 | 49.5 | 76.9 | 95.0 | |
| GPT-5.5 gpt | 89.50 | 87.50 | 100.00 | 90.00 | 80.00 pass | 88.60 | 78.9 | 70.5 | 100.0 | |
| 豆包 Pro doubao | 85.30 | 92.50 | 100.00 | 80.00 | 70.00 pass | 88.54 | 100.0 | 64.4 | 100.0 | |
| GPT-6.1 Sol gpt | 89.50 | 86.80 | 100.00 | 85.00 | 90.00 pass | 88.29 | 97.7 | 67.8 | 100.0 | |
| GPT-6 Astra gpt | 89.50 | 84.50 | 100.00 | 85.00 | 80.00 pass | 87.25 | 41.1 | 65.4 | 100.0 | |
| GPT-6 Sol gpt | 88.70 | 84.50 | 97.20 | 90.00 | 90.00 pass | 86.81 | 97.6 | 73.4 | 100.0 | |
| Grok 4 grok | 77.80 | 89.80 | 32.80 | 70.00 | 55.00 warn | 83.20 | 84.1 | 25.6 | 100.0 | |
| Claude Sonnet 4.6 claude | 69.50 | 89.80 | 85.50 | 90.00 | 90.00 pass | 78.64 | 88.0 | 50.3 | 100.0 | |
| Gemini 2.5 Pro gemini | 65.30 | 89.80 | 75.00 | 65.00 | 90.00 pass | 76.33 | 96.8 | 38.4 | 95.0 | |
| Gemini 3.1 Pro gemini | 64.50 | 89.80 | 100.00 | 60.00 | 90.00 pass | 75.89 | 87.6 | 37.4 | 100.0 | |
| GPT-6 Luna gpt | 64.50 | 86.80 | 100.00 | 90.00 | 80.00 pass | 74.54 | 100.0 | 51.6 | 100.0 | |
| GLM-4.6 zhipu | 19.50 | 89.80 | 50.00 | 50.00 | 40.00 warn | 51.14 | 91.5 | 73.3 | 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。