额外格式 失败
看看哪些 AI 模型最常遇到 额外格式，让你在选择前先发现稳定性风险。
41/41
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 | 额外格式 次数 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #43 | Claude Opus 4.6 medium | Anthropic | 5 | 7.7 | $3.059 | 13/22 | 34.3s |
| #62 | Claude Sonnet 4.6 none | Anthropic | 4 | 7.3 | $0.661 | 12/22 | 8.12s |
| #108 | Claude Sonnet 5 none | Anthropic | 4 | 6.3 | $0.548 | 8/22 | 6.04s |
| #154 | KAT-Coder-Air V2.5 low | Kwaipilot | 4 | 5.4 | $0.041 | 7/22 | 10.1s |
| #40 | Claude Sonnet 4.6 medium | Anthropic | 3 | 7.8 | $2.057 | 14/22 | 25.9s |
| #48 | Grok Build 0.1 medium | X AI | 3 | 7.6 | $1.097 | 14/22 | 52.1s |
| #65 | Claude Opus 4.8 none | Anthropic | 3 | 7.3 | $1.166 | 13/22 | 4.91s |
| #83 | MiMo-V2.5-Pro medium | Xiaomi | 3 | 6.9 | $0.187 | 12/22 | 33.9s |
| #140 | KAT-Coder-Air V2.5 high | Kwaipilot | 3 | 5.6 | $0.077 | 7/22 | 15.9s |
| #178 | KAT-Coder-Air V2.5 none | Kwaipilot | 3 | 4.8 | $0.067 | 5/22 | 12.2s |
| #98 | MiMo-V2.5 medium | Xiaomi | 2 | 6.5 | $0.082 | 12/22 | 32.2s |
| #133 | North Mini Code medium | Cohere | 2 | 5.9 | $0.000 | 9/22 | 137.1s |
| #146 | DeepSeek V4 Flash none | DeepSeek | 2 | 5.6 | $0.044 | 5/22 | 36.8s |
| #167 | North Mini Code none | Cohere | 2 | 5.1 | $0.000 | 4/22 | 29.9s |
| #169 | DeepSeek V3.2 none | DeepSeek | 2 | 5.0 | $0.054 | 6/22 | 18.3s |