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。
197/197
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） |
|---|---|---|---|---|---|---|---|
| #6 | GPT-5.6 Sol medium | OpenAI | 1 | 4.7 | $1.316 | 0/1 | 4.94s |
| #7 | GPT-5.6 Sol high | OpenAI | 1 | 4.7 | $1.234 | 0/1 | 6.97s |
| #8 | GPT-5.5 low | OpenAI | 1 | 3.0 | $1.253 | 0/1 | 10.1s |
| #12 | Qwen3.7 Max medium | Qwen | 1 | 3.0 | $1.116 | 0/1 | 33.4s |
| #14 | Qwen3.8 Max medium | Qwen | 1 | 3.0 | $0.728 | 0/1 | 11.5s |
| #15 | GPT-5.5 medium | OpenAI | 1 | 2.8 | $4.137 | 0/1 | 37.9s |
| #18 | GPT-5.3-Codex medium | OpenAI | 1 | 2.8 | $0.920 | 0/1 | 14.4s |
| #20 | Muse Spark 1.2 high | Meta | 1 | 3.0 | $1.632 | 0/1 | 15.0s |
| #21 | Gemini 3.5 Flash Lite high | 1 | 3.0 | $0.538 | 0/1 | 9.68s | |
| #23 | Claude Opus 4.7 medium | Anthropic | 1 | 3.0 | $1.477 | 0/1 | 2.25s |
| #24 | Qwen3.8 Max low | Qwen | 1 | 3.0 | $0.687 | 0/1 | 12.4s |
| #25 | Muse Spark 1.2 medium | Meta | 1 | 3.0 | $1.227 | 0/1 | 12.8s |
| #26 | Muse Spark 1.1 medium | Meta | 1 | 3.0 | $1.357 | 0/1 | 18.5s |
| #28 | Claude Opus 5 low | Anthropic | 1 | 3.0 | $1.121 | 0/1 | 6.48s |
| #29 | GPT-5.4 medium | OpenAI | 1 | 3.0 | $1.533 | 0/1 | 14.0s |