YZ Index
评测数据
当前展示:Run #373 | 2026-10-12 | 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 |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-6 Astra gpt | 97.00 | 70.30 | 100.00 | 79.70 | 80.00 pass | 84.99 | 40.4 | 61.5 | 100.0 | |
| GPT-6.1 Sol gpt | 97.00 | 70.30 | 100.00 | 79.70 | 80.00 pass | 84.99 | 97.3 | 61.5 | 100.0 | |
| 豆包 Pro doubao | 91.50 | 76.30 | 100.00 | 80.00 | 65.00 pass | 84.66 | 100.0 | 47.3 | 100.0 | |
| GPT-6 Sol gpt | 94.80 | 68.10 | 100.00 | 79.70 | 80.00 pass | 82.79 | 97.1 | 61.4 | 100.0 | |
| GPT-6 Luna gpt | 97.00 | 56.00 | 86.40 | 85.00 | 70.00 pass | 78.55 | 100.0 | 45.2 | 100.0 | |
| Claude Sonnet 4.6 claude | 72.00 | 76.30 | 100.00 | 80.00 | 70.00 pass | 73.94 | 85.1 | 46.3 | 100.0 | |
| DeepSeek V4 Pro deepseek | 72.00 | 73.30 | 100.00 | 59.70 | 80.00 pass | 72.59 | 97.6 | 64.2 | 90.0 | |
| GPT-5.5 gpt | 72.00 | 73.30 | 100.00 | 90.00 | 80.00 pass | 72.59 | 72.6 | 48.4 | 100.0 | |
| Gemini 3.1 Pro gemini | 72.00 | 71.10 | 95.80 | 60.00 | 80.00 pass | 71.60 | 84.6 | 37.2 | 100.0 | |
| GPT-o3 gpt | 69.80 | 73.30 | 100.00 | 65.00 | 80.00 pass | 71.38 | 41.6 | 47.7 | 95.0 | |
| Gemini 2.5 Pro gemini | 69.50 | 73.30 | 75.00 | 80.00 | 80.00 pass | 71.21 | 95.5 | 60.1 | 90.0 | |
| Grok 4 grok | 63.70 | 73.30 | 76.40 | 75.00 | 65.00 pass | 68.02 | 80.9 | 34.7 | 100.0 | |
| Claude Opus 4.7 claude | 47.00 | 73.30 | 100.00 | 90.00 | 90.00 pass | 58.84 | 21.8 | 37.5 | 100.0 | |
| GLM-4.6 zhipu | 16.70 | 71.10 | 25.00 | 25.00 | 65.00 pass | 41.18 | 85.8 | 41.8 | 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。