AI BENCHY 失败分析
答案错误 失败
看看哪些 AI 模型最常遇到 答案错误，让你在选择前先发现稳定性风险。 排序方式: 测试正确 ↑.
| 排名 | 模型 | 公司 | 答案错误 次数 | 分数 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|
| #93 | Qwen3.6 Plus Preview medium | Qwen | 2 | 6.3 | 9/19 | 15.2s |
| #76 | Kimi K2.5 medium | Moonshot AI | 5 | 6.8 | 10/21 | 98.4s |
| #78 | Qwen3.6 27B medium | Qwen | 6 | 6.8 | 10/21 | 59.7s |
| #80 | Mimo V2 Omni medium | Xiaomi | 5 | 6.7 | 10/21 | 41.2s |
| #81 | Mercury 2 medium | Inception | 8 | 6.6 | 10/21 | 2.24s |
| #85 | Gemma 4 31B none | 8 | 6.5 | 10/21 | 4.05s | |
| #87 | Gemini 3.1 Flash Lite minimal | 8 | 6.4 | 10/21 | 1.33s | |
| #88 | Qwen3.7 Plus none | Qwen | 10 | 6.4 | 10/21 | 2.85s |
| #89 | Hy3 preview low | Tencent | 4 | 6.4 | 10/21 | 24.6s |
| #91 | GPT-5.5 none | OpenAI | 11 | 6.4 | 10/21 | 1.89s |
| #83 | Step 3.5 Flash none | Stepfun | 1 | 6.6 | 6/12 | 39.0s |
| #59 | GLM 5V Turbo medium | Z.ai | 7 | 7.2 | 11/21 | 23.1s |
| #66 | Qwen3.5-35B-A3B medium | Qwen | 2 | 7.1 | 11/21 | 72.6s |
| #67 | MiniMax M3 medium | Minimax | 3 | 7.1 | 11/21 | 68.2s |
| #70 | GPT-5.4 Nano medium | OpenAI | 8 | 7.0 | 11/21 | 12.0s |