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: 总成本 ↑.
215/215
筛选模型
没有模型匹配当前搜索和筛选条件。
| 排名 | 模型 | 公司 | 答案错误 次数 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #49 | DeepSeek V4 Flash high | DeepSeek | 6 | 7.7 | $0.041 | 13/22 | 49.7s |
| #147 | GLM 5 none | Z.ai | 12 | 5.7 | $0.041 | 9/21 | 4.03s |
| #186 | GPT-5.4 Nano none | OpenAI | 15 | 4.8 | $0.041 | 4/22 | 2.57s |
| #164 | KAT-Coder-Air V2.5 low | Kwaipilot | 7 | 5.4 | $0.041 | 7/22 | 10.1s |
| #156 | DeepSeek V4 Flash none | DeepSeek | 12 | 5.6 | $0.042 | 5/22 | 36.8s |
| #119 | MiMo-V2-Flash medium | Xiaomi | 5 | 6.3 | $0.043 | 12/21 | 20.1s |
| #117 | LongCat 2.0 none | Meituan | 14 | 6.3 | $0.044 | 7/22 | 5.18s |
| #153 | Mimo V2 PRO none | Xiaomi | 11 | 5.6 | $0.045 | 7/21 | 2.27s |
| #128 | Gemini 3.1 Flash Lite none | 11 | 6.1 | $0.046 | 9/22 | 1.75s | |
| #176 | GLM 5 Turbo none | Z.ai | 13 | 5.1 | $0.047 | 6/21 | 2.82s |
| #126 | Gemini 3.1 Flash Lite minimal | 8 | 6.1 | $0.047 | 10/22 | 1.86s | |
| #141 | Hy3 preview high | Tencent | 3 | 5.9 | $0.048 | 11/21 | 56.6s |
| #155 | KAT-Coder-Air V2.5 medium | Kwaipilot | 11 | 5.6 | $0.048 | 8/22 | 8.42s |
| #151 | GLM 5V Turbo none | Z.ai | 11 | 5.6 | $0.052 | 8/21 | 2.99s |
| #112 | Gemini 3.1 Flash Lite Preview none | 7 | 6.4 | $0.052 | 12/22 | 1.58s |