YZ Index
评测数据
当前展示:Run #438 | 2026-11-22 | 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 | 88.50 | 77.40 | 42.80 | 93.80 | 65.00 pass | 83.51 | 85.8 | 32.2 | 100.0 | |
| GPT-o3 gpt | 71.60 | 86.70 | 75.00 | 100.00 | 65.00 pass | 78.40 | 43.1 | 35.7 | 95.0 | |
| Claude Opus 4.7 claude | 99.10 | 52.40 | 33.30 | 50.00 | 90.00 pass | 78.09 | 23.2 | 21.5 | 100.0 | |
| 豆包 Pro doubao | 73.80 | 72.40 | 70.50 | 96.30 | 90.00 pass | 73.17 | 100.0 | 45.7 | 95.0 | |
| GPT-6.1 Sol gpt | 74.30 | 66.70 | 75.00 | 94.70 | 80.00 pass | 70.88 | 94.8 | 35.7 | 100.0 | |
| GPT-6 Luna gpt | 71.40 | 61.90 | 63.60 | 75.00 | 80.00 pass | 67.13 | 100.0 | 36.0 | 100.0 | |
| GPT-6 Astra gpt | 74.30 | 57.60 | 75.00 | 94.70 | 80.00 pass | 66.79 | 33.3 | 29.3 | 100.0 | |
| GPT-6 Sol gpt | 74.30 | 52.80 | 75.00 | 91.00 | 90.00 pass | 64.63 | 93.6 | 35.2 | 100.0 | |
| DeepSeek V4 Pro deepseek | 50.00 | 77.40 | 75.00 | 100.00 | 90.00 pass | 62.33 | 96.4 | 45.4 | 90.0 | |
| Claude Sonnet 4.6 claude | 48.60 | 77.40 | 50.00 | 56.30 | 80.00 pass | 61.56 | 80.3 | 20.4 | 100.0 | |
| Gemini 2.5 Pro gemini | 50.00 | 66.70 | 75.00 | 83.80 | 90.00 pass | 57.52 | 92.7 | 45.6 | 90.0 | |
| Gemini 3.1 Pro gemini | 50.00 | 66.70 | 75.00 | 100.00 | 80.00 pass | 57.52 | 78.7 | 27.0 | 100.0 | |
| GPT-5.5 gpt | 49.30 | 60.10 | 75.00 | 100.00 | 80.00 pass | 54.16 | 61.4 | 32.5 | 95.0 | |
| GLM-4.6 zhipu | 25.00 | 66.70 | 75.00 | 100.00 | 50.00 warn | 43.77 | 89.3 | 22.5 | 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。