YZ Index
评测数据
当前展示:Run #437 | 2026-11-21 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| 豆包 Pro doubao | 98.80 | 81.70 | 100.00 | 76.70 | 90.00 pass | 91.11 | 100.0 | 62.3 | 100.0 | |
| Grok 4 grok | 98.80 | 58.30 | 55.70 | 68.30 | 80.00 pass | 80.58 | 84.8 | 38.5 | 100.0 | |
| GPT-6 Luna gpt | 99.30 | 56.70 | 100.00 | 71.30 | 80.00 pass | 80.13 | 100.0 | 46.5 | 100.0 | |
| GPT-o3 gpt | 98.80 | 51.70 | 87.50 | 77.90 | 70.00 pass | 77.61 | 42.8 | 53.9 | 100.0 | |
| Gemini 3.1 Pro gemini | 75.00 | 70.00 | 100.00 | 51.70 | 70.00 pass | 72.75 | 84.9 | 34.0 | 100.0 | |
| Claude Opus 4.7 claude | 74.30 | 70.00 | 87.50 | 60.40 | 80.00 pass | 72.37 | 24.0 | 35.4 | 100.0 | |
| GPT-6.1 Sol gpt | 74.30 | 70.00 | 100.00 | 77.50 | 80.00 pass | 72.37 | 95.8 | 43.3 | 100.0 | |
| Gemini 2.5 Pro gemini | 75.00 | 63.30 | 61.40 | 52.90 | 65.00 pass | 69.74 | 94.3 | 28.5 | 90.0 | |
| GPT-6 Astra gpt | 74.30 | 63.30 | 100.00 | 72.50 | 80.00 pass | 69.35 | 35.4 | 35.6 | 100.0 | |
| GPT-6 Sol gpt | 74.30 | 63.30 | 100.00 | 77.50 | 80.00 pass | 69.35 | 95.3 | 37.4 | 100.0 | |
| Claude Sonnet 4.6 claude | 73.80 | 63.30 | 100.00 | 46.30 | 80.00 pass | 69.08 | 82.8 | 31.0 | 100.0 | |
| DeepSeek V4 Pro deepseek | 75.00 | 58.30 | 94.50 | 60.40 | 65.00 pass | 67.49 | 96.8 | 43.9 | 90.0 | |
| GPT-5.5 gpt | 49.30 | 63.40 | 100.00 | 81.70 | 80.00 pass | 55.65 | 66.4 | 33.8 | 100.0 | |
| GLM-4.6 zhipu | 25.00 | 70.00 | 25.00 | 58.80 | 40.00 warn | 45.25 | 87.7 | 54.4 | 55.0 | |
| 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。