谜题求解：API 错误
谜题求解
API 错误
看看哪些 AI 模型在 谜题求解 上最容易遇到 API 错误，更快找出薄弱点。 排序方式: 测试正确 ↑.
11/11
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 | API 错误 次数 | 分类得分 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #192 | Laguna M.1 none | Poolside | 1 | 3.0 | $0.009 | 0/3 | 891ms |
| #207 | Nemotron 3 Nano Omni 30b A3b Reasoning medium | NVIDIA | 1 | 2.9 | $0.000 | 0/3 | 1.40s |
| #208 | Nemotron 3 Nano Omni 30b A3b Reasoning none | NVIDIA | 1 | 3.0 | $0.000 | 0/3 | 532ms |
| #210 | LFM2-24B-A2B none | Liquid | 1 | 3.8 | $0.001 | 0/3 | 1.78s |
| #153 | Hy3 preview low | Tencent | 1 | 5.3 | $0.015 | 1/3 | 7.51s |
| #175 | Qwen3.6 Plus Preview medium | Qwen | 2 | 5.3 | $0.000 | 1/3 | 7.52s |
| #186 | Laguna M.1 medium | Poolside | 1 | 5.3 | $0.033 | 1/3 | 10.2s |
| #198 | Laguna Xs.2 medium | Poolside | 1 | 5.3 | $0.015 | 1/3 | 1.93s |
| #205 | Laguna Xs.2 none | Poolside | 1 | 5.3 | $0.004 | 1/3 | 650ms |
| #135 | Hy3 preview high | Tencent | 1 | 7.7 | $0.048 | 2/3 | 27.9s |
| #173 | DeepSeek V3.2 none | DeepSeek | 1 | 7.6 | $0.054 | 2/3 | 6.91s |