AI BENCHY 失败分析
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: 分数 ↑.
| 排名 | 模型 | 公司 | 额外格式 次数 | 分数 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|
| #163 | Granite 4.1 8B none | IBM Granite | 1 | 4.0 | 2/21 | 728ms |
| #161 | Qwen3.5-9B medium | Qwen | 1 | 4.2 | 3/21 | 82.2s |
| #156 | Hy3 preview none | Tencent | 1 | 4.4 | 4/21 | 12.9s |
| #152 | MiMo-V2-Flash none | Xiaomi | 1 | 4.6 | 4/21 | 2.76s |
| #143 | MiMo-V2.5 none | Xiaomi | 1 | 4.9 | 5/21 | 2.20s |
| #140 | Qwen3 Coder Next none | Qwen | 1 | 4.9 | 5/21 | 8.62s |
| #139 | DeepSeek V4 Flash none | DeepSeek | 2 | 5.0 | 5/21 | 26.8s |
| #133 | DeepSeek V3.2 none | DeepSeek | 2 | 5.2 | 6/21 | 13.8s |
| #127 | Grok 4.20 none | X AI | 1 | 5.4 | 6/18 | 1.11s |
| #121 | Owl Alpha none | Openrouter | 1 | 5.5 | 7/21 | 9.88s |
| #113 | DeepSeek V4 Pro none | DeepSeek | 1 | 5.7 | 7/21 | 12.4s |
| #101 | Mimo V2 Omni none | Xiaomi | 1 | 6.0 | 8/21 | 2.44s |
| #84 | Grok 4.20 Multi Agent Beta medium | X AI | 2 | 6.6 | 8/18 | 9.69s |
| #80 | Mimo V2 Omni medium | Xiaomi | 1 | 6.7 | 10/21 | 41.2s |
| #79 | Hunter Alpha medium | OpenRouter | 1 | 6.7 | 8/18 | 10.3s |