YZ Index
评测数据
当前展示:Run #415 | 2026-11-07 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-6 Astra gpt | 99.80 | 84.80 | 75.00 | 85.00 | 70.00 pass | 93.05 | 41.5 | 46.0 | 100.0 | |
| GPT-6 Luna gpt | 99.80 | 71.30 | 70.50 | 90.00 | 70.00 pass | 86.98 | 100.0 | 48.9 | 100.0 | |
| GPT-6 Sol gpt | 99.80 | 69.50 | 75.00 | 90.00 | 70.00 pass | 86.17 | 97.0 | 42.8 | 100.0 | |
| GPT-6.1 Sol gpt | 99.80 | 69.50 | 75.00 | 90.00 | 70.00 pass | 86.17 | 97.0 | 42.8 | 100.0 | |
| GPT-o3 gpt | 91.60 | 72.30 | 75.00 | 86.30 | 80.00 pass | 82.92 | 45.2 | 48.3 | 100.0 | |
| GPT-5.5 gpt | 99.80 | 61.30 | 75.00 | 90.00 | 80.00 pass | 82.48 | 75.4 | 45.2 | 100.0 | |
| 豆包 Pro doubao | 96.80 | 58.50 | 63.60 | 95.00 | 90.00 pass | 79.57 | 100.0 | 42.3 | 100.0 | |
| Gemini 3.1 Pro gemini | 75.00 | 84.50 | 50.00 | 90.00 | 80.00 pass | 79.28 | 73.2 | 30.0 | 100.0 | |
| Claude Opus 4.7 claude | 70.00 | 89.80 | 50.00 | 76.30 | 90.00 pass | 78.91 | 28.1 | 32.4 | 100.0 | |
| Claude Sonnet 4.6 claude | 89.30 | 59.30 | 45.70 | 51.30 | 70.00 pass | 75.80 | 83.6 | 39.4 | 100.0 | |
| Grok 4 grok | 71.40 | 74.50 | 19.40 | 87.50 | 70.00 pass | 72.80 | 84.1 | 31.9 | 100.0 | |
| Gemini 2.5 Pro gemini | 50.00 | 81.80 | 75.00 | 85.00 | 90.00 pass | 64.31 | 89.3 | 50.8 | 90.0 | |
| GLM-4.6 zhipu | 50.00 | 81.80 | 50.00 | 25.00 | 80.00 pass | 64.31 | 87.8 | 44.4 | 70.0 | |
| DeepSeek V4 Pro deepseek | 50.00 | 79.50 | 75.00 | 75.00 | 80.00 pass | 63.28 | 91.2 | 47.7 | 85.0 | |
| 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。