AI BENCHY 失败分析
超时 失败
看看哪些 AI 模型最常遇到 超时，让你在选择前先发现稳定性风险。 排序方式: 失败次数 ↑.
| 排名 | 模型 | 公司 | 超时 次数 | 分数 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|
| #3 | Claude Opus 4.7 medium | Anthropic | 1 | 9.2 | 16/18 | 3.53s |
| #10 | Qwen3.5-27B medium | Qwen | 1 | 8.4 | 13/18 | 53.0s |
| #13 | GLM 5 medium | Z.ai | 1 | 8.4 | 13/18 | 23.3s |
| #14 | Gemma 4 31B medium | 1 | 8.3 | 13/18 | 24.9s | |
| #18 | GLM 5 Turbo medium | Z.ai | 1 | 8.1 | 12/18 | 17.7s |
| #23 | MiMo-V2-Pro medium | Xiaomi | 1 | 8.1 | 12/18 | 12.3s |
| #26 | Claude Sonnet 4.6 medium | Anthropic | 1 | 8.0 | 13/18 | 12.7s |
| #40 | GPT-5.2 medium | OpenAI | 1 | 7.5 | 11/18 | 14.0s |
| #41 | MiMo-V2-Flash medium | Xiaomi | 1 | 7.5 | 11/18 | 23.4s |
| #45 | GPT-5 Mini medium | OpenAI | 1 | 7.0 | 9/18 | 24.0s |
| #51 | Nemotron 3 Super medium | NVIDIA | 1 | 6.7 | 9/18 | 19.1s |
| #52 | Grok 4.1 Fast medium | X AI | 1 | 6.7 | 9/18 | 23.9s |
| #57 | GPT-5 Nano medium | OpenAI | 1 | 6.3 | 7/18 | 44.1s |
| #60 | Gemma 4 26B A4B none | 1 | 6.2 | 7/18 | 6.59s | |
| #92 | Qwen3 Coder Next medium | Qwen | 1 | 4.7 | 3/18 | 10.8s |