编程：答案错误
编程
答案错误
看看哪些 AI 模型在 编程 上最容易遇到 答案错误，更快找出薄弱点。
197/197
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 | 答案错误 次数 | 分类得分 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #139 | KAT-Coder-Pro V2.5 none | Kwaipilot | 2 | 6.1 | $0.487 | 1/3 | 22.5s |
| #142 | GLM 5V Turbo medium | Z.ai | 2 | 6.0 | $0.457 | 1/3 | 63.4s |
| #146 | Qwen3.5-27B none | Qwen | 2 | 5.8 | $0.058 | 1/3 | 1.80s |
| #147 | Gemini 3.1 Flash Lite Preview low | 2 | 5.5 | $0.646 | 1/3 | 1.39s | |
| #149 | Gemini 3.1 Flash Lite low | 2 | 5.5 | $0.621 | 1/3 | 1.53s | |
| #153 | Laguna XS 2.1 medium | Poolside | 2 | 5.5 | $0.068 | 1/3 | 70.3s |
| #155 | Gemini 3.1 Flash Lite Preview none | 2 | 5.5 | $0.052 | 1/3 | 967ms | |
| #159 | LongCat 2.0 none | Meituan | 2 | 5.5 | $0.044 | 1/3 | 2.85s |
| #160 | MiMo-V2.5 medium | Xiaomi | 2 | 6.2 | $0.104 | 1/3 | 97.1s |
| #161 | Ling-3.0-flash high | Inclusionai | 2 | 6.7 | $0.021 | 1/3 | 14.9s |
| #165 | Seed-2.0-Lite none | Bytedance Seed | 2 | 5.6 | $0.066 | 1/3 | 2.83s |
| #167 | GPT-5.6 Luna low | OpenAI | 2 | 5.5 | $0.051 | 1/3 | 4.61s |
| #169 | Gemini 3.1 Flash Lite minimal | 2 | 5.5 | $0.047 | 1/3 | 831ms | |
| #171 | gpt-oss-120b medium | OpenAI | 2 | 5.9 | $0.019 | 1/3 | 38.4s |
| #172 | Qwen3.7 Flash none | Qwen | 2 | 5.5 | $0.019 | 1/3 | 1.55s |