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 ↑.
87/87
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） |
|---|---|---|---|---|---|---|---|
| #135 | Claude Haiku 5.5 medium | Anthropic | 2 | 0.8 | $0.063 | 13/23 | 7.71s |
| #158 | GLM 5.3 Flash high | Z.ai | 2 | 0.7 | $0.056 | 13/23 | 29.1s |
| #175 | Claude Opus 5 none | Anthropic | 2 | 0.7 | $2.766 | 12/23 | 10.8s |
| #179 | MiMo-V2.5 medium | Xiaomi | 2 | 0.7 | $0.130 | 12/23 | 49.2s |
| #257 | DeepSeek V4 Flash 0423 none | DeepSeek | 2 | 0.6 | $0.149 | 5/23 | 38.6s |
| #258 | Grok 4.20 medium | X AI | 2 | 0.6 | $1.246 | 11/23 | 36.7s |
| #277 | MiMo-V2.6-Pro none | Xiaomi | 2 | 0.6 | $0.193 | 9/23 | 9.66s |
| #313 | Seed-2.0-Code none | Bytedance Seed | 2 | 0.5 | $0.257 | 6/23 | 17.5s |
| #316 | North Mini Code medium | Cohere | 2 | 0.5 | $0.000 | 8/23 | 140.4s |
| #360 | North Mini Code default | Cohere | 2 | 0.5 | $0.000 | 4/23 | 29.5s |
| #364 | DeepSeek V3.2 none | DeepSeek | 2 | 0.5 | $0.119 | 6/23 | 22.4s |
| #391 | Grok 4.20 Multi Agent Beta medium | X AI | 2 | 0.4 | $5.599 | 8/18 | 9.69s |
| #67 | MiMo-V2.6-Pro medium | Xiaomi | 3 | 0.9 | $0.586 | 14/23 | 148.1s |
| #195 | Claude Opus 4.8 none | Anthropic | 3 | 0.7 | $2.334 | 13/23 | 7.54s |
| #204 | Grok Build 0.1 medium | X AI | 3 | 0.7 | $1.419 | 13/23 | 55.1s |