YZ Index
评测数据
当前展示:Run #425 | 2026-11-13 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.7 claude | 99.50 | 89.30 | 65.00 | 43.80 | 90.00 pass | 94.91 | 27.8 | 28.5 | 100.0 | |
| GPT-6 Astra gpt | 94.80 | 89.30 | 75.00 | 100.00 | 80.00 pass | 92.33 | 42.7 | 50.0 | 100.0 | |
| Claude Sonnet 4.6 claude | 99.50 | 61.30 | 60.80 | 37.50 | 80.00 pass | 82.31 | 84.5 | 25.7 | 100.0 | |
| GPT-5.5 gpt | 100.00 | 59.00 | 75.00 | 93.80 | 80.00 pass | 81.55 | 75.0 | 37.9 | 100.0 | |
| GPT-o3 gpt | 94.90 | 61.30 | 75.00 | 75.00 | 80.00 pass | 79.78 | 43.5 | 36.8 | 95.0 | |
| GPT-6.1 Sol gpt | 94.80 | 59.00 | 75.00 | 100.00 | 80.00 pass | 78.69 | 96.6 | 37.9 | 100.0 | |
| GPT-6 Sol gpt | 99.50 | 37.00 | 75.00 | 96.30 | 90.00 pass | 71.38 | 95.4 | 28.7 | 100.0 | |
| Grok 4 grok | 73.30 | 61.30 | 33.30 | 50.00 | 80.00 pass | 67.90 | 78.2 | 21.3 | 100.0 | |
| GPT-6 Luna gpt | 93.60 | 19.80 | 75.00 | 88.80 | 80.00 pass | 60.39 | 100.0 | 23.9 | 100.0 | |
| 豆包 Pro doubao | 75.00 | 36.30 | 75.00 | 77.50 | 90.00 pass | 57.59 | 100.0 | 27.1 | 95.0 | |
| Gemini 3.1 Pro gemini | 75.00 | 36.30 | 75.00 | 56.30 | 80.00 pass | 57.59 | 76.3 | 15.6 | 100.0 | |
| Gemini 2.5 Pro gemini | 47.80 | 61.30 | 50.00 | 61.30 | 90.00 pass | 53.88 | 89.6 | 27.5 | 85.0 | |
| DeepSeek V4 Pro deepseek | 50.00 | 34.00 | 75.00 | 100.00 | 80.00 pass | 42.80 | 91.1 | 26.2 | 90.0 | |
| GLM-4.6 zhipu | 25.00 | 47.00 | 50.00 | - | 50.00 warn | 34.90 | 85.7 | 17.9 | 56.3 | |
| 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。