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: 测试正确 ↑.
364/364
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） |
|---|---|---|---|---|---|---|---|
| #10 | GPT-6 Astra low | OpenAI | 3.0 | 9.5 | $1.289 | 0/1 | 4.12s |
| #15 | GPT-5.6 Sol medium | OpenAI | 4.7 | 9.4 | $0.508 | 0/1 | 4.94s |
| #16 | GPT-6 Astra medium | OpenAI | 2.8 | 9.4 | $1.774 | 0/1 | 5.98s |
| #17 | GPT-5.6 Sol high | OpenAI | 4.7 | 9.4 | $0.446 | 0/1 | 6.97s |
| #18 | GPT-5.5 low | OpenAI | 3.0 | 9.3 | $1.257 | 0/1 | 10.1s |
| #19 | Grok 4.6 high | X AI | 3.0 | 9.3 | $1.908 | 0/1 | 54.2s |
| #20 | Claude Opus 5.5 low | Anthropic | 3.0 | 9.3 | $0.952 | 0/1 | 27.9s |
| #21 | Claude Opus 5.5 high | Anthropic | 3.0 | 9.3 | $1.266 | 0/1 | 27.4s |
| #24 | Claude Opus 5 high | Anthropic | 3.0 | 9.2 | $3.070 | 0/1 | 14.1s |
| #25 | Claude Opus 5 medium | Anthropic | 3.0 | 9.2 | $1.586 | 0/1 | 7.03s |
| #27 | Qwen3.8 Max (0902) low | Qwen | 3.0 | 9.1 | $0.664 | 0/1 | 13.3s |
| #28 | Seed 2.1 Turbo low | Bytedance Seed | 3.0 | 9.1 | $1.546 | 0/1 | 171.9s |
| #29 | Qwen3.7 Max medium | Qwen | 3.0 | 9.1 | $1.157 | 0/1 | 33.4s |
| #30 | Claude Opus 5.5 medium | Anthropic | 3.0 | 9.0 | $1.080 | 0/1 | 27.7s |
| #31 | Claude Fable 5.1 high | Anthropic | 3.0 | 9.0 | $3.364 | 0/1 | 30.6s |