YZ Index
评测数据
当前展示:Run #382 | 2026-10-17 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-o3 gpt | 100.00 | 87.80 | 100.00 | 81.70 | 70.00 pass | 94.51 | 49.6 | 64.6 | 100.0 | |
| Grok 4 grok | 91.70 | 85.00 | 86.10 | 85.80 | 70.00 pass | 88.69 | 89.2 | 57.2 | 100.0 | |
| Gemini 3.1 Pro gemini | 100.00 | 72.80 | 75.00 | 81.70 | 80.00 pass | 87.76 | 89.3 | 37.7 | 100.0 | |
| GPT-6 Sol gpt | 100.00 | 71.80 | 100.00 | 72.20 | 90.00 pass | 87.31 | 97.4 | 56.1 | 100.0 | |
| DeepSeek V4 Pro deepseek | 100.00 | 70.00 | 100.00 | 81.70 | 70.00 pass | 86.50 | 98.8 | 49.0 | 100.0 | |
| Claude Opus 4.7 claude | 75.00 | 95.00 | 80.60 | 66.70 | 100.00 pass | 84.00 | 27.3 | 33.5 | 100.0 | |
| GPT-6.1 Sol gpt | 83.30 | 81.80 | 100.00 | 67.20 | 80.00 pass | 82.63 | 96.8 | 54.8 | 100.0 | |
| GPT-6 Luna gpt | 70.80 | 91.80 | 100.00 | 77.50 | 80.00 pass | 80.25 | 100.0 | 55.9 | 100.0 | |
| GPT-5.5 gpt | 75.00 | 85.00 | 100.00 | 81.70 | 80.00 pass | 79.50 | 75.7 | 46.0 | 100.0 | |
| GPT-6 Astra gpt | 83.30 | 71.80 | 100.00 | 67.20 | 80.00 pass | 78.13 | 37.6 | 48.0 | 100.0 | |
| Gemini 2.5 Pro gemini | 94.50 | 55.00 | 100.00 | 81.70 | 80.00 pass | 76.73 | 97.1 | 42.6 | 100.0 | |
| Claude Sonnet 4.6 claude | 75.00 | 77.00 | 85.50 | 65.80 | 90.00 pass | 75.90 | 86.0 | 32.1 | 100.0 | |
| 豆包 Pro doubao | 83.30 | 51.60 | 100.00 | 86.70 | 70.00 pass | 69.04 | 100.0 | 40.1 | 100.0 | |
| GLM-4.6 zhipu | 41.70 | 87.80 | 50.00 | 12.50 | 40.00 warn | 62.45 | 92.7 | 61.0 | 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。