YZ Index
评测数据
当前显示:Run #362 | 2026-10-05 | 154题题库 | 公式 v7 | 判定 v6.4
数据公开说明:为防止题库污染和过拟合,题目原文与期望答案不公开。本页面展示模型回答、评分、判定方式等透明数据。完整方法论请参见方法论页面。
| 模型 | 代码执行 | 资料约束 | 工学判断 | 任务表达 | 誠実性 | 主评分 | 性价比 | 安定性 | 可用性 | 详细操作 |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-6 Sol gpt | 92.10 | 70.00 | 86.10 | 86.10 | 90.00 pass | 82.16 | 39.4 | 44.5 | 100.0 | |
| Claude Opus 4.7 claude | 81.50 | 81.50 | 90.00 | 72.80 | 93.30 pass | 81.50 | 5.6 | 39.2 | 100.0 | |
| GPT-o3 gpt | 84.40 | 75.90 | 79.70 | 78.30 | 80.00 pass | 80.58 | 9.5 | 38.1 | 98.0 | |
| GPT-5.5 gpt | 87.60 | 71.80 | 90.00 | 90.80 | 83.30 pass | 80.49 | 19.5 | 43.7 | 100.0 | |
| Grok 4 grok | 85.30 | 74.50 | 70.20 | 85.80 | 88.30 pass | 80.44 | 25.8 | 37.4 | 100.0 | |
| GPT-6 Astra gpt | 90.50 | 68.10 | 89.60 | 86.10 | 90.00 pass | 80.42 | 8.2 | 43.0 | 100.0 | |
| Claude Sonnet 4.6 claude | 86.60 | 72.10 | 80.80 | 75.40 | 90.00 pass | 80.08 | 26.4 | 38.2 | 100.0 | |
| GPT-6 Luna gpt | 81.70 | 76.60 | 86.20 | 88.80 | 81.70 pass | 79.41 | 100.0 | 39.5 | 100.0 | |
| GPT-6.1 Sol gpt | 90.50 | 64.00 | 89.00 | 86.10 | 90.00 pass | 78.58 | 38.4 | 39.6 | 100.0 | |
| 豆包 Pro doubao | 86.60 | 64.60 | 87.80 | 93.30 | 78.30 pass | 76.70 | 94.4 | 38.2 | 98.9 | |
| DeepSeek V4 Pro deepseek | 75.10 | 77.90 | 81.70 | 66.10 | 85.00 pass | 76.36 | 44.6 | 42.7 | 90.5 | |
| Gemini 3.1 Pro gemini | 73.80 | 74.60 | 70.00 | 75.80 | 86.70 pass | 74.16 | 24.8 | 24.9 | 100.0 | |
| Qwen3 Max qwen | 76.50 | 70.00 | 42.30 | 50.30 | 81.70 pass | 73.58 | 47.4 | 24.6 | 100.0 | |
| Gemini 2.5 Pro gemini | 75.20 | 68.60 | 68.30 | 65.80 | 96.70 pass | 72.23 | 37.0 | 30.6 | 94.0 | |
| GLM-4.6 zhipu | 49.50 | 70.10 | 25.00 | 25.00 | 43.30 warn | 58.77 | 28.6 | 56.4 | 52.0 |
API访问:如需通过程序访问评测数据,请使用
API 。。