常识问答：答案错误
常识问答
答案错误
看看哪些 AI 模型在 常识问答 上最容易遇到 答案错误，更快找出薄弱点。
323/323
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 | 分类得分 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #277 | KAT-Coder-Pro V2.5 medium | Kwaipilot | 1 | 0.3 | $0.669 | 0/1 | 9.87s |
| #278 | GLM 5 Turbo medium | Z.ai | 1 | 0.3 | $0.323 | 0/1 | 40.2s |
| #279 | DeepSeek V4 Flash 0731 none | DeepSeek | 1 | 0.3 | $0.053 | 0/1 | 1.85s |
| #280 | Gemini 3.1 Flash Lite Preview none | 1 | 0.3 | $0.090 | 0/1 | 814ms | |
| #281 | Claude Fable 5.1 low | Anthropic | 1 | 0.3 | $3.395 | 0/1 | 19.2s |
| #282 | Gemini 3.5 Flash none | 1 | 0.3 | $1.752 | 0/1 | 4.87s | |
| #283 | Beam 1.0 501B A23B high | Reflection | 1 | 0.3 | $0.674 | 0/1 | 16.4s |
| #284 | Trinity Large Thinking medium | Arcee AI | 1 | 0.3 | $0.862 | 0/1 | 97.4s |
| #285 | Apodex 1.1 Mini low | Apodex | 1 | 0.3 | $0.000 | 0/1 | 54.7s |
| #286 | Inkling none | Thinkingmachines | 1 | 0.3 | $0.292 | 0/1 | 670ms |
| #287 | Qwen3.5 Plus 2026-04-20 none | Qwen | 1 | 0.3 | $0.216 | 0/1 | 33.3s |
| #289 | Granite 4.2 8B medium | IBM Granite | 1 | 0.3 | $0.029 | 0/1 | 2.24s |
| #290 | KAT-Coder-Pro V2.5 default | Kwaipilot | 1 | 0.3 | $0.742 | 0/1 | 9.91s |
| #291 | Apodex 1.1 Mini medium | Apodex | 1 | 0.3 | $0.000 | 0/1 | 42.9s |
| #292 | Claude Sonnet 4.6 none | Anthropic | 1 | 0.3 | $0.661 | 0/1 | 4.67s |