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。 排序方式: 失败次数 ↑.
42/42
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） |
|---|---|---|---|---|---|---|---|
| #33 | Kimi K3 max | Moonshot AI | 1 | 8.0 | $3.112 | 16/22 | 122.5s |
| #41 | Claude Opus 4.8 low | Anthropic | 1 | 7.8 | $2.077 | 16/22 | 12.7s |
| #46 | DeepSeek V4 Pro high | DeepSeek | 1 | 7.7 | $0.200 | 10/22 | 79.1s |
| #47 | MiniMax M3 medium | Minimax | 1 | 7.6 | $0.286 | 12/22 | 75.0s |
| #58 | Qwen3.5-27B medium | Qwen | 1 | 7.4 | $1.627 | 13/22 | 111.9s |
| #73 | Grok 4.3 medium | X AI | 1 | 7.1 | $0.779 | 13/22 | 47.4s |
| #74 | GLM 5.1 medium | Z.ai | 1 | 7.1 | $0.535 | 13/22 | 46.8s |
| #75 | Grok 4.20 medium | X AI | 1 | 7.1 | $0.777 | 12/22 | 29.5s |
| #82 | DeepSeek V4 Pro none | DeepSeek | 1 | 6.9 | $0.096 | 10/22 | 11.6s |
| #109 | Mimo V2 PRO medium | Xiaomi | 1 | 6.3 | $0.333 | 12/21 | 22.2s |
| #111 | LongCat 2.0 none | Meituan | 1 | 6.3 | $0.044 | 7/22 | 5.18s |
| #113 | MiMo-V2-Flash medium | Xiaomi | 1 | 6.3 | $0.043 | 12/21 | 20.1s |
| #134 | Mimo V2 Omni medium | Xiaomi | 1 | 5.9 | $0.683 | 10/21 | 41.2s |
| #148 | Owl Alpha none | Openrouter | 1 | 5.6 | $0.000 | 7/21 | 9.88s |
| #149 | KAT-Coder-Air V2.5 medium | Kwaipilot | 1 | 5.6 | $0.048 | 8/22 | 8.42s |