YZ Index
评测数据
当前展示:Run #354 | 2026-10-01 | 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 | 94.30 | 97.00 | 100.00 | 86.70 | 75.00 pass | 95.52 | 100.0 | 54.6 | 100.0 | |
| GPT-6 Sol gpt | 95.00 | 82.90 | 100.00 | 72.20 | 90.00 pass | 89.56 | 97.6 | 66.7 | 100.0 | |
| GPT-6 Astra gpt | 95.00 | 80.70 | 100.00 | 67.20 | 100.00 pass | 88.57 | 41.5 | 63.9 | 100.0 | |
| GPT-6.1 Sol gpt | 95.00 | 80.70 | 100.00 | 72.20 | 100.00 pass | 88.57 | 97.7 | 68.6 | 100.0 | |
| Grok 4 grok | 100.00 | 74.20 | 56.90 | 77.80 | 90.00 pass | 88.39 | 88.2 | 45.4 | 100.0 | |
| Claude Sonnet 4.6 claude | 95.00 | 79.20 | 94.50 | 79.70 | 90.00 pass | 87.89 | 89.7 | 66.5 | 100.0 | |
| Claude Opus 4.7 claude | 75.00 | 100.00 | 100.00 | 74.70 | 100.00 pass | 86.25 | 28.8 | 51.4 | 100.0 | |
| GLM-4.6 zhipu | 75.00 | 95.00 | 25.00 | 41.70 | 90.00 pass | 84.00 | 97.7 | 73.3 | 70.0 | |
| DeepSeek V4 Pro deepseek | 70.00 | 95.00 | 94.50 | 77.20 | 90.00 pass | 81.25 | 98.7 | 70.0 | 95.0 | |
| Gemini 3.1 Pro gemini | 70.00 | 95.00 | 100.00 | 76.70 | 90.00 pass | 81.25 | 89.6 | 48.3 | 100.0 | |
| GPT-o3 gpt | 75.00 | 80.30 | 95.80 | 81.70 | 90.00 pass | 77.39 | 45.3 | 49.9 | 100.0 | |
| Gemini 2.5 Pro gemini | 70.00 | 85.90 | 64.60 | 86.70 | 100.00 pass | 77.16 | 97.2 | 71.2 | 90.0 | |
| GPT-6 Luna gpt | 70.00 | 85.90 | 94.50 | 77.50 | 90.00 pass | 77.16 | 100.0 | 49.2 | 100.0 | |
| Qwen3 Max qwen | 70.00 | 85.90 | 36.80 | 75.60 | 80.00 pass | 77.16 | 99.2 | 37.9 | 100.0 | |
| GPT-5.5 gpt | 50.00 | 95.00 | 100.00 | 81.70 | 100.00 pass | 70.25 | 74.2 | 38.7 | 100.0 |
API 访问:如需程序化访问评测数据,请使用我们的 API。