YZ Index
评测数据
当前展示:Run #359 | 2026-10-04 | 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 | 97.00 | 62.70 | 69.50 | 43.90 | 90.00 pass | 81.57 | 25.2 | 31.7 | 100.0 | |
| GPT-o3 gpt | 97.80 | 57.00 | 75.00 | 75.00 | 90.00 pass | 79.44 | 44.3 | 44.8 | 95.0 | |
| 豆包 Pro doubao | 96.00 | 50.60 | 75.00 | 100.00 | 70.00 pass | 75.57 | 100.0 | 43.9 | 100.0 | |
| Grok 4 grok | 95.30 | 48.40 | 50.00 | 75.00 | 60.00 pass | 74.20 | 81.7 | 28.3 | 100.0 | |
| DeepSeek V4 Pro deepseek | 75.00 | 71.70 | 100.00 | 75.00 | 80.00 pass | 73.52 | 97.8 | 64.9 | 90.0 | |
| Gemini 3.1 Pro gemini | 75.00 | 69.50 | 50.00 | 75.00 | 80.00 pass | 72.53 | 82.8 | 22.4 | 100.0 | |
| Qwen3 Max qwen | 61.60 | 72.90 | 50.00 | 43.90 | 90.00 pass | 66.69 | 98.1 | 27.9 | 100.0 | |
| GPT-6 Luna gpt | 69.80 | 57.50 | 75.00 | 85.00 | 90.00 pass | 64.27 | 100.0 | 29.9 | 100.0 | |
| Gemini 2.5 Pro gemini | 69.80 | 56.20 | 75.00 | 60.00 | 90.00 pass | 63.68 | 94.3 | 41.7 | 90.0 | |
| GPT-6.1 Sol gpt | 69.80 | 53.60 | 75.00 | 100.00 | 90.00 pass | 62.51 | 93.9 | 28.3 | 100.0 | |
| Claude Sonnet 4.6 claude | 71.90 | 50.60 | 75.00 | 83.90 | 90.00 pass | 62.32 | 79.9 | 35.3 | 100.0 | |
| GPT-6 Sol gpt | 69.80 | 50.60 | 100.00 | 94.70 | 90.00 pass | 61.16 | 94.2 | 44.6 | 100.0 | |
| GPT-5.5 gpt | 75.00 | 41.90 | 75.00 | 100.00 | 100.00 pass | 60.11 | 66.6 | 26.2 | 100.0 | |
| GPT-6 Astra gpt | 44.80 | 62.70 | 75.00 | 94.70 | 90.00 pass | 52.86 | 29.4 | 27.8 | 100.0 | |
| GLM-4.6 zhipu | 50.00 | 14.30 | 75.00 | 25.00 | 50.00 warn | 33.94 | 85.4 | 73.0 | 45.0 |
API 访问:如需程序化访问评测数据,请使用我们的 API。