YZ Index
评测数据
当前展示:Run #435 | 2026-11-19 | 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 Sol gpt | 68.70 | 95.00 | 75.00 | 86.30 | 90.00 pass | 80.54 | 96.6 | 39.1 | 100.0 | |
| GPT-5.5 gpt | 74.30 | 83.90 | 50.00 | 90.00 | 80.00 pass | 78.62 | 71.6 | 28.9 | 100.0 | |
| GPT-o3 gpt | 73.80 | 83.30 | 75.00 | 85.00 | 90.00 pass | 78.08 | 43.5 | 35.3 | 100.0 | |
| Claude Sonnet 4.6 claude | 68.50 | 88.30 | 75.00 | 38.80 | 80.00 pass | 77.41 | 83.9 | 28.9 | 100.0 | |
| Claude Opus 4.7 claude | 57.60 | 95.00 | 75.00 | 71.30 | 100.00 pass | 74.43 | 25.9 | 32.0 | 100.0 | |
| 豆包 Pro doubao | 48.80 | 100.00 | 75.00 | 85.00 | 100.00 pass | 71.84 | 100.0 | 29.8 | 100.0 | |
| Gemini 3.1 Pro gemini | 47.30 | 95.00 | 50.00 | 85.00 | 80.00 pass | 68.77 | 81.2 | 25.1 | 100.0 | |
| GPT-6 Astra gpt | 49.30 | 83.90 | 75.00 | 85.00 | 80.00 pass | 64.87 | 31.8 | 27.5 | 100.0 | |
| GPT-6.1 Sol gpt | 49.30 | 83.90 | 75.00 | 86.30 | 80.00 pass | 64.87 | 92.9 | 29.6 | 100.0 | |
| GPT-6 Luna gpt | 38.20 | 95.00 | 75.00 | 86.30 | 80.00 pass | 63.76 | 100.0 | 29.6 | 100.0 | |
| GLM-4.6 zhipu | 16.70 | 95.00 | 25.00 | 70.00 | 100.00 pass | 51.94 | 94.4 | 47.2 | 70.0 | |
| Grok 4 grok | 23.50 | 83.30 | 30.60 | 86.30 | 80.00 pass | 50.41 | 73.2 | 26.6 | 100.0 | |
| DeepSeek V4 Pro deepseek | 16.70 | 83.90 | 50.00 | 95.00 | 75.00 pass | 46.94 | 94.0 | 33.7 | 80.0 | |
| Gemini 2.5 Pro gemini | 0.00 | 83.90 | 50.00 | 91.30 | 90.00 pass | 37.76 | 88.1 | 26.1 | 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。