常识问答 排名
看看哪些 AI 模型在 常识问答 上表现最好，哪些更稳定，以及差距主要出现在哪里。 排序方式: 指标 ↑.
461/461
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 | 常识问答 得分 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #16 | GPT-6.1 Sol low | OpenAI | 0.3 | 1.0 | $0.460 | 0/1 | 3.80s |
| #22 | GPT-6.1 Terra high | OpenAI | 0.3 | 0.9 | $0.396 | 0/1 | 7.53s |
| #23 | GPT-6 Astra medium | OpenAI | 0.3 | 0.9 | $2.707 | 0/1 | 5.98s |
| #27 | GPT-6.1 Terra max | OpenAI | 0.3 | 0.9 | $2.209 | 0/1 | 156.2s |
| #38 | GPT-5.5 medium | OpenAI | 0.3 | 0.9 | $4.809 | 0/1 | 37.9s |
| #43 | GPT-5.3-Codex medium | OpenAI | 0.3 | 0.9 | $1.318 | 0/1 | 14.4s |
| #47 | Pareto default | Unbiased | 0.3 | 0.9 | $1.520 | 0/1 | 32.1s |
| #64 | Pareto 26.9 default | Unbiased | 0.3 | 0.9 | $1.864 | 0/1 | 28.6s |
| #75 | MiMo-V2.6-Pro medium | Xiaomi | 0.3 | 0.9 | $0.586 | 0/1 | 527.1s |
| #133 | Muse Wonder max | Meta | 0.3 | 0.8 | $4.766 | 0/1 | 41.3s |
| #143 | Muse Wonder high | Meta | 0.3 | 0.8 | $3.426 | 0/1 | 28.8s |
| #151 | Muse Wonder xhigh | Meta | 0.3 | 0.8 | $4.012 | 0/1 | 25.4s |
| #168 | Inkling medium | Thinkingmachines | 0.3 | 0.7 | $0.527 | 0/1 | 21.7s |
| #171 | Claude Fable 5.1 medium | Anthropic | 0.3 | 0.7 | $3.197 | 0/1 | 24.3s |
| #176 | Gemini 3.5 Flash Lite medium | 0.3 | 0.7 | $0.440 | 0/1 | 3.75s |