YZ Index
评测数据
当前展示:Run #439 | 2026-11-23 | 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 | 96.00 | 57.60 | 75.00 | 68.80 | 90.00 pass | 78.72 | 25.2 | 29.6 | 100.0 | |
| GPT-6.1 Sol gpt | 99.30 | 52.60 | 100.00 | 90.00 | 80.00 pass | 78.29 | 96.5 | 55.2 | 100.0 | |
| Claude Sonnet 4.6 claude | 98.60 | 52.60 | 62.50 | 76.30 | 80.00 pass | 77.90 | 84.7 | 39.5 | 100.0 | |
| GPT-6 Sol gpt | 99.30 | 45.90 | 100.00 | 85.00 | 80.00 pass | 75.27 | 96.0 | 45.1 | 100.0 | |
| GPT-5.5 gpt | 74.30 | 57.60 | 100.00 | 90.00 | 90.00 pass | 66.79 | 71.7 | 54.3 | 95.0 | |
| Grok 4 grok | 77.60 | 52.60 | 56.00 | 95.00 | 45.00 warn | 66.35 | 79.5 | 41.8 | 100.0 | |
| GPT-6 Astra gpt | 74.30 | 52.60 | 100.00 | 79.70 | 80.00 pass | 64.54 | 34.2 | 41.6 | 100.0 | |
| GPT-6 Luna gpt | 96.40 | 25.00 | 95.50 | 90.00 | 80.00 pass | 64.27 | 100.0 | 36.7 | 100.0 | |
| 豆包 Pro doubao | 73.80 | 47.60 | 88.60 | 91.30 | 70.00 pass | 62.01 | 100.0 | 54.0 | 95.0 | |
| GPT-o3 gpt | 73.80 | 39.30 | 100.00 | 91.30 | 90.00 pass | 58.28 | 39.2 | 46.4 | 95.0 | |
| DeepSeek V4 Pro deepseek | 50.00 | 64.30 | 100.00 | 84.70 | 70.00 pass | 56.44 | 96.3 | 55.4 | 90.0 | |
| GLM-4.6 zhipu | 50.00 | 64.30 | 25.00 | 25.00 | 45.00 warn | 56.44 | 89.8 | 51.6 | 55.0 | |
| Gemini 3.1 Pro gemini | 50.00 | 52.60 | 100.00 | 85.00 | 80.00 pass | 51.17 | 79.7 | 32.3 | 100.0 | |
| Gemini 2.5 Pro gemini | 50.00 | 45.90 | 94.50 | 86.30 | 80.00 pass | 48.16 | 92.5 | 47.3 | 90.0 | |
| 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。