编程 排名
看看哪些 AI 模型在 编程 上表现最好，哪些更稳定，以及差距主要出现在哪里。 排序方式: 响应时间（平均） ↑.
408/408
筛选模型
没有模型匹配当前搜索和筛选条件。
| 排名 | 模型 | 公司 | 编程 得分 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #20 | GPT-6 Astra medium | OpenAI | 1.0 | 0.9 | $2.707 | 3/3 | 4.95s |
| #210 | GLM 5.1 none | Z.ai | 0.4 | 0.6 | $0.451 | 0/3 | 4.96s |
| #282 | GLM 5 none | Z.ai | 0.4 | 0.6 | $0.140 | 0/3 | 5.12s |
| #12 | GPT-6.1 Sol high | OpenAI | 1.0 | 1.0 | $0.755 | 3/3 | 5.18s |
| #254 | Claude Sonnet 4.6 none | Anthropic | 0.5 | 0.6 | $0.661 | 1/3 | 5.19s |
| #255 | GLM 5.3 FlashX low | Z.ai | 0.5 | 0.6 | $0.157 | 1/3 | 5.45s |
| #159 | Claude Opus 5 none | Anthropic | 0.6 | 0.7 | $2.766 | 1/3 | 5.54s |
| #331 | Qwen3.5-9B none | Qwen | 0.4 | 0.5 | $0.041 | 0/3 | 5.60s |
| #178 | Claude Sonnet 5.5 low | Anthropic | 0.3 | 0.7 | $0.909 | 0/3 | 5.81s |
| #7 | GPT-6 Astra high | OpenAI | 1.0 | 1.0 | $4.458 | 3/3 | 5.86s |
| #172 | Gemini 3 Flash Preview low | 0.6 | 0.7 | $0.254 | 1/3 | 6.00s | |
| #205 | Claude Sonnet 5.5 medium | Anthropic | 0.3 | 0.7 | $1.100 | 0/3 | 6.07s |
| #321 | Dots 3 Note Preview default | Dots Studio | 0.5 | 0.5 | $0.000 | 0/3 | 6.12s |
| #207 | Mercury 2.5 Preview high | Inception | 0.7 | 0.6 | $0.040 | 1/3 | 6.13s |
| #25 | Gemini 3.7 Flash low | 0.8 | 0.9 | $0.183 | 2/3 | 6.20s |