AI BENCHY 分类
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 ↑.
失败原因
| 排名 | 模型 | 公司 | 常识问答 得分 | 分数 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|
| #3 | Claude Opus 4.7 medium | Anthropic | 3.0 | 8.9 | 0/1 | 2.25s |
| #4 | GPT-5.5 medium | OpenAI | 2.8 | 8.9 | 0/1 | 37.9s |
| #5 | Claude Opus 4.7 none | Anthropic | 3.0 | 8.9 | 0/1 | 1.46s |
| #6 | GPT-5.5 low | OpenAI | 3.0 | 8.9 | 0/1 | 10.1s |
| #9 | Qwen3.6 Max Preview medium | Qwen | 3.0 | 8.5 | 0/1 | 60.6s |
| #10 | Gemini 3 PRO Preview medium | 0.0 | 8.4 | 0/0 | 0ms | |
| #11 | Seed-2.0-Lite medium | Bytedance Seed | 3.0 | 8.3 | 0/1 | 48.3s |
| #12 | Qwen3.5 Plus 2026-02-15 medium | Qwen | 3.0 | 8.2 | 0/1 | 103.8s |
| #13 | GPT-5.3-Codex medium | OpenAI | 2.8 | 8.2 | 0/1 | 14.4s |
| #14 | Gemma 4 31B medium | 3.0 | 8.2 | 0/1 | 90.1s | |
| #15 | Qwen3.6 Plus Preview medium | Qwen | 0.0 | 8.2 | 0/0 | 0ms |
| #17 | Qwen3.5-27B medium | Qwen | 3.0 | 8.1 | 0/1 | 85.1s |
| #19 | GLM 5 medium | Z.ai | 3.0 | 8.1 | 0/1 | 67.4s |
| #20 | GLM 5 Turbo medium | Z.ai | 3.0 | 8.1 | 0/1 | 40.2s |
| #21 | Qwen3.6 35B A3B medium | Qwen | 3.0 | 8.0 | 0/1 | 32.9s |