AI BENCHY 失败分析
API 错误 失败
看看哪些 AI 模型最常遇到 API 错误，让你在选择前先发现稳定性风险。 排序方式: 响应时间（平均） ↓.
| 排名 | 模型 | 公司 | API 错误 次数 | 分数 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|
| #105 | Nemotron 3 Super medium | NVIDIA | 3 | 5.8 | 8/21 | 32.0s |
| #26 | Qwen3.6 Plus medium | Qwen | 1 | 7.9 | 14/21 | 30.7s |
| #100 | Grok Build 0.1 none | X AI | 3 | 6.0 | 7/19 | 28.7s |
| #43 | MiMo-V2.5-Pro medium | Xiaomi | 1 | 7.5 | 12/21 | 26.1s |
| #89 | Hy3 preview low | Tencent | 7 | 6.4 | 10/21 | 24.6s |
| #51 | Mimo V2 PRO medium | Xiaomi | 1 | 7.4 | 12/21 | 22.2s |
| #126 | gpt-oss-120b none | OpenAI | 3 | 5.4 | 6/19 | 21.6s |
| #64 | MiMo-V2-Flash medium | Xiaomi | 1 | 7.2 | 12/21 | 20.1s |
| #46 | Qwen3.6 35B A3B medium | Qwen | 2 | 7.4 | 13/21 | 18.1s |
| #149 | Nemotron 3 Nano Omni 30b A3b Reasoning medium | NVIDIA | 6 | 4.6 | 4/19 | 17.1s |
| #33 | Hy3 preview medium | Tencent | 3 | 7.7 | 14/21 | 16.3s |
| #93 | Qwen3.6 Plus Preview medium | Qwen | 8 | 6.3 | 9/19 | 15.2s |
| #41 | Nemotron 3 Ultra 550b A55b medium | NVIDIA | 1 | 7.5 | 13/21 | 15.1s |
| #92 | Laguna M.1 medium | Poolside | 4 | 6.4 | 9/19 | 14.7s |
| #133 | DeepSeek V3.2 none | DeepSeek | 4 | 5.2 | 6/21 | 13.8s |