YZ Index
评测数据
当前展示:Run #394 | 2026-10-25 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| Grok 4 grok | 93.90 | 100.00 | 62.00 | 71.30 | 70.00 pass | 96.65 | 89.7 | 51.1 | 100.0 | |
| GPT-5.5 gpt | 100.00 | 89.30 | 100.00 | 90.00 | 100.00 pass | 95.19 | 81.6 | 75.1 | 100.0 | |
| GPT-6 Luna gpt | 98.00 | 89.30 | 95.50 | 83.80 | 80.00 pass | 94.09 | 100.0 | 67.2 | 100.0 | |
| 豆包 Pro doubao | 96.90 | 89.30 | 95.50 | 87.50 | 70.00 pass | 93.48 | 100.0 | 68.3 | 100.0 | |
| GPT-6 Sol gpt | 92.50 | 84.00 | 100.00 | 84.70 | 100.00 pass | 88.68 | 98.0 | 71.6 | 100.0 | |
| GPT-6 Astra gpt | 94.80 | 74.00 | 100.00 | 84.70 | 80.00 pass | 85.44 | 41.0 | 65.3 | 100.0 | |
| Claude Opus 4.7 claude | 75.00 | 89.30 | 58.30 | 57.50 | 90.00 pass | 81.44 | 24.7 | 27.8 | 100.0 | |
| Gemini 3.1 Pro gemini | 75.00 | 89.30 | 100.00 | 90.00 | 90.00 pass | 81.44 | 89.1 | 49.9 | 100.0 | |
| GPT-6.1 Sol gpt | 94.80 | 61.50 | 100.00 | 84.70 | 80.00 pass | 79.82 | 96.9 | 62.4 | 100.0 | |
| GPT-o3 gpt | 70.50 | 84.50 | 91.70 | 86.30 | 100.00 pass | 76.80 | 44.8 | 50.7 | 100.0 | |
| Claude Sonnet 4.6 claude | 50.00 | 100.00 | 31.80 | 70.00 | 90.00 pass | 72.50 | 80.4 | 18.1 | 100.0 | |
| Gemini 2.5 Pro gemini | 47.80 | 100.00 | 93.20 | 82.50 | 100.00 pass | 71.29 | 96.4 | 79.7 | 90.0 | |
| DeepSeek V4 Pro deepseek | 50.00 | 89.30 | 100.00 | 84.70 | 70.00 pass | 67.69 | 97.0 | 63.8 | 90.0 | |
| GLM-4.6 zhipu | 25.00 | 39.30 | 50.00 | 65.00 | 90.00 pass | 31.44 | 80.1 | 41.2 | 65.0 | |
| 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。