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） |
|---|---|---|---|---|---|---|
| #61 | Seed-2.0-Lite none | Bytedance Seed | 10 | 6.2 | 8/18 | 2.53s |
| #44 | GPT-5.4 Mini medium | OpenAI | 4 | 7.3 | 9/18 | 15.2s |
| #45 | GPT-5 Mini medium | OpenAI | 4 | 7.0 | 9/18 | 24.0s |
| #46 | Kimi K2.5 medium | Moonshot AI | 4 | 7.0 | 9/18 | 72.4s |
| #47 | Grok 4.20 medium | X AI | 3 | 7.0 | 9/18 | 10.3s |
| #49 | Qwen3.5 Plus 2026-02-15 none | Qwen | 9 | 6.8 | 9/18 | 2.60s |
| #51 | Nemotron 3 Super medium | NVIDIA | 3 | 6.7 | 9/18 | 19.1s |
| #52 | Grok 4.1 Fast medium | X AI | 3 | 6.7 | 9/18 | 23.9s |
| #53 | GLM 5 none | Z.ai | 9 | 6.6 | 9/18 | 4.23s |
| #43 | Qwen3.5-35B-A3B medium | Qwen | 2 | 7.4 | 10/18 | 44.5s |
| #48 | Gemma 4 31B none | 5 | 6.9 | 10/18 | 4.02s | |
| #31 | GLM 5V Turbo medium | Z.ai | 3 | 7.8 | 11/18 | 15.0s |
| #32 | Qwen3.5-Flash medium | Qwen | 1 | 7.8 | 11/18 | 66.7s |
| #34 | Kimi K2.6 medium | Moonshot AI | 2 | 7.7 | 11/18 | 45.2s |
| #35 | MiMo-V2-Omni medium | Xiaomi | 3 | 7.7 | 11/18 | 16.8s |