YZ Index
评测数据
当前展示:Run #364 | 2026-10-07 | 128 题库 | 公式 v7 | 判分 v6.4
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 | 100.00 | 94.40 | 94.50 | 81.70 | 90.00 pass | 97.48 | 30.5 | 71.5 | 100.0 | |
| Qwen3 Max qwen | 97.50 | 92.20 | 69.50 | 51.70 | 75.00 pass | 95.12 | 99.7 | 30.4 | 100.0 | |
| GPT-6.1 Sol gpt | 100.00 | 80.70 | 100.00 | 77.50 | 80.00 pass | 91.32 | 97.7 | 54.9 | 100.0 | |
| GPT-o3 gpt | 100.00 | 75.10 | 100.00 | 81.70 | 65.00 pass | 88.80 | 47.4 | 43.7 | 100.0 | |
| GPT-5.5 gpt | 100.00 | 74.30 | 100.00 | 77.50 | 80.00 pass | 88.44 | 77.8 | 57.9 | 100.0 | |
| Grok 4 grok | 96.80 | 68.30 | 69.50 | 95.80 | 65.00 pass | 83.98 | 87.3 | 37.5 | 100.0 | |
| GPT-6 Luna gpt | 75.00 | 94.30 | 81.40 | 81.70 | 80.00 pass | 83.69 | 100.0 | 47.9 | 100.0 | |
| GPT-6 Sol gpt | 100.00 | 55.00 | 94.80 | 77.50 | 80.00 pass | 79.75 | 96.3 | 46.2 | 100.0 | |
| GPT-6 Astra gpt | 75.00 | 80.70 | 97.80 | 72.50 | 80.00 pass | 77.57 | 37.9 | 39.6 | 100.0 | |
| Gemini 3.1 Pro gemini | 78.60 | 75.00 | 100.00 | 76.70 | 80.00 pass | 76.98 | 87.8 | 38.7 | 100.0 | |
| Claude Sonnet 4.6 claude | 75.00 | 62.70 | 100.00 | 90.80 | 80.00 pass | 69.47 | 85.3 | 45.9 | 100.0 | |
| 豆包 Pro doubao | 74.30 | 51.50 | 100.00 | 86.70 | 80.00 pass | 64.04 | 100.0 | 36.9 | 100.0 | |
| DeepSeek V4 Pro deepseek | 50.00 | 75.00 | 100.00 | 81.70 | 80.00 pass | 61.25 | 97.2 | 47.6 | 90.0 | |
| GLM-4.6 zhipu | 25.00 | 100.00 | 100.00 | 48.30 | 50.00 warn | 58.75 | 95.7 | 40.7 | 75.0 | |
| Gemini 2.5 Pro gemini | 52.10 | 62.80 | 75.00 | 81.70 | 90.00 pass | 56.92 | 94.2 | 38.6 | 90.0 |
API 访问:如需程序化访问评测数据,请使用我们的 API。