AI BENCHY 分类失败
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 ↑.
| 排名 | 模型 | 公司 | API 错误 次数 | 分类得分 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|
| #145 | Laguna M.1 none | Poolside | 1 | 3.0 | 0/3 | 891ms |
| #149 | Nemotron 3 Nano Omni 30b A3b Reasoning medium | NVIDIA | 1 | 2.9 | 0/3 | 1.40s |
| #160 | LFM2-24B-A2B none | Liquid | 1 | 3.8 | 0/3 | 1.78s |
| #162 | Nemotron 3 Nano Omni 30b A3b Reasoning none | NVIDIA | 1 | 3.0 | 0/3 | 532ms |
| #89 | Hy3 preview low | Tencent | 1 | 5.3 | 1/3 | 7.51s |
| #92 | Laguna M.1 medium | Poolside | 1 | 5.3 | 1/3 | 10.2s |
| #93 | Qwen3.6 Plus Preview medium | Qwen | 2 | 5.3 | 1/3 | 7.52s |
| #103 | DeepSeek V4 Pro high | DeepSeek | 1 | 5.9 | 1/3 | 34.8s |
| #107 | Laguna Xs.2 medium | Poolside | 1 | 5.3 | 1/3 | 1.93s |
| #146 | Laguna Xs.2 none | Poolside | 1 | 5.3 | 1/3 | 650ms |
| #82 | Hy3 preview high | Tencent | 1 | 7.7 | 2/3 | 27.9s |
| #133 | DeepSeek V3.2 none | DeepSeek | 1 | 7.6 | 2/3 | 6.91s |