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） |
|---|---|---|---|---|---|---|---|
| #310 | Trinity Large Thinking high | Arcee AI | 1 | 0.3 | $0.794 | 0/1 | 125.1s |
| #311 | GPT-5 Nano medium | OpenAI | 1 | 0.3 | $0.159 | 0/1 | 20.1s |
| #312 | GPT-5.4 Mini none | OpenAI | 1 | 0.3 | $0.170 | 0/1 | 1.33s |
| #313 | Mistral Large 4 none | Mistral | 1 | 0.3 | $0.290 | 0/1 | 780ms |
| #314 | Seed 2.1 Turbo none | Bytedance Seed | 1 | 0.3 | $0.182 | 0/1 | 2.12s |
| #315 | Dots 3 Note Preview medium | Dots Studio | 1 | 0.3 | $0.000 | 0/1 | 15.5s |
| #316 | Gemma 4 31B none | 1 | 0.3 | $0.031 | 0/1 | 1.25s | |
| #317 | Space Bunny Alpha low | Stealth | 1 | 0.3 | $0.000 | 0/1 | 340.9s |
| #318 | GPT-5.4 none | OpenAI | 1 | 0.3 | $0.690 | 0/1 | 990ms |
| #319 | Solar Pro 4 none | Upstage | 1 | 0.3 | $0.022 | 0/1 | 1.29s |
| #321 | GLM 5 none | Z.ai | 1 | 0.3 | $0.140 | 0/1 | 3.62s |
| #322 | Qwen3.5-35B-A3B none | Qwen | 1 | 0.3 | $0.120 | 0/1 | 493ms |
| #323 | Nemotron 3.5 Super VL none | NVIDIA | 1 | 0.3 | $0.040 | 0/1 | 10.6s |
| #324 | Nemotron 3 Super medium | NVIDIA | 1 | 0.3 | $0.081 | 0/1 | 55.3s |
| #325 | Kimi K2.6 none | Moonshot AI | 1 | 0.3 | $0.486 | 0/1 | 1.36s |