谜题求解 排名
看看哪些 AI 模型在 谜题求解 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: 总成本 ↓.
216/216
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 | 谜题求解 得分 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #55 | Nemotron 3 Ultra medium | NVIDIA | 5.5 | 7.5 | $0.774 | 1/3 | 3.54s |
| #60 | GPT-5.4 Mini medium | OpenAI | 7.8 | 7.5 | $0.756 | 2/3 | 4.37s |
| #137 | Grok 4.20 Beta medium | X AI | 10.0 | 6.0 | $0.750 | 3/3 | 3.52s |
| #94 | Qwen3.6 35B A3B medium | Qwen | 8.0 | 6.7 | $0.746 | 2/3 | 5.95s |
| #3 | Gemini 3 Flash Preview medium | 10.0 | 9.6 | $0.742 | 3/3 | 4.05s | |
| #56 | Kimi K2.7 Code medium | Moonshot AI | 5.9 | 7.5 | $0.740 | 1/3 | 41.0s |
| #89 | Qwen3.6 Flash medium | Qwen | 8.2 | 6.9 | $0.738 | 2/3 | 6.29s |
| #140 | Mimo V2 Omni medium | Xiaomi | 5.9 | 5.9 | $0.683 | 1/3 | 2.38s |
| #43 | GPT-5.6 Terra medium | OpenAI | 8.4 | 7.8 | $0.676 | 2/3 | 3.78s |
| #67 | Claude Sonnet 4.6 none | Anthropic | 7.7 | 7.3 | $0.661 | 2/3 | 2.53s |
| #27 | Muse Spark 1.1 low | Meta | 8.3 | 8.3 | $0.647 | 2/3 | 6.60s |
| #110 | Gemini 3.1 Flash Lite Preview low | 10.0 | 6.5 | $0.646 | 3/3 | 1.69s | |
| #28 | Gemini 2.5 Flash medium | 7.7 | 8.2 | $0.643 | 2/3 | 3.18s | |
| #12 | Gemini 3.5 Flash medium | 7.7 | 9.1 | $0.642 | 2/3 | 2.38s | |
| #111 | Gemini 3.1 Flash Lite low | 10.0 | 6.5 | $0.621 | 3/3 | 1.40s |