YZ Index
评测数据
当前展示:Run #381 | 2026-10-16 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| 豆包 Pro doubao | 95.30 | 72.60 | 100.00 | 100.00 | 90.00 pass | 85.09 | 100.0 | 68.6 | 100.0 | |
| Grok 4 grok | 95.30 | 63.10 | 66.70 | 85.00 | 90.00 pass | 80.81 | 86.9 | 40.2 | 100.0 | |
| DeepSeek V4 Pro deepseek | 72.00 | 88.10 | 100.00 | 75.00 | 90.00 pass | 79.25 | 98.5 | 77.2 | 90.0 | |
| GPT-6 Astra gpt | 72.00 | 88.10 | 100.00 | 100.00 | 90.00 pass | 79.25 | 39.5 | 53.3 | 100.0 | |
| GPT-6.1 Sol gpt | 72.00 | 88.10 | 100.00 | 100.00 | 90.00 pass | 79.25 | 97.0 | 53.3 | 100.0 | |
| Claude Opus 4.7 claude | 68.90 | 79.00 | 100.00 | 63.90 | 100.00 pass | 73.45 | 25.0 | 39.1 | 100.0 | |
| GPT-6 Luna gpt | 69.80 | 73.60 | 100.00 | 100.00 | 90.00 pass | 71.51 | 100.0 | 49.0 | 100.0 | |
| Gemini 3.1 Pro gemini | 54.50 | 88.10 | 75.00 | 75.00 | 90.00 pass | 69.62 | 86.5 | 24.6 | 100.0 | |
| GPT-5.5 gpt | 47.00 | 81.40 | 100.00 | 100.00 | 90.00 pass | 62.48 | 68.7 | 37.2 | 100.0 | |
| Gemini 2.5 Pro gemini | 50.00 | 77.60 | 75.00 | 75.00 | 90.00 pass | 62.42 | 93.7 | 48.2 | 85.0 | |
| GPT-o3 gpt | 38.30 | 88.10 | 100.00 | 75.00 | 80.00 pass | 60.71 | 38.7 | 36.5 | 95.0 | |
| Claude Sonnet 4.6 claude | 44.80 | 77.60 | 100.00 | 83.90 | 90.00 pass | 59.56 | 80.7 | 38.8 | 100.0 | |
| GPT-6 Sol gpt | 44.80 | 63.10 | 100.00 | 100.00 | 90.00 pass | 53.04 | 92.1 | 29.5 | 100.0 | |
| GLM-4.6 zhipu | 25.00 | 28.90 | 25.00 | 25.00 | 90.00 pass | 26.76 | 78.3 | 34.8 | 50.0 | |
| 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。