工具调用 排名
看看哪些 AI 模型在 工具调用 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 ↓.
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 | 工具调用 得分 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #136 | Step 3.5 Flash medium | Stepfun | 10.0 | 6.0 | $0.108 | 1/1 | 11.9s |
| #138 | GPT-5.6 Terra none | OpenAI | 9.6 | 6.0 | $0.349 | 1/1 | 3.10s |
| #139 | Gemini 3 PRO Preview medium | 10.0 | 6.0 | $0.385 | 1/1 | 12.0s | |
| #140 | Mimo V2 Omni medium | Xiaomi | 10.0 | 5.9 | $0.683 | 1/1 | 14.0s |
| #141 | Hy3 preview high | Tencent | 10.0 | 5.9 | $0.048 | 1/1 | 78.8s |
| #143 | North Mini Code medium | Cohere | 10.0 | 5.9 | $0.000 | 1/1 | 3.93s |
| #144 | Kimi K2.6 none | Moonshot AI | 10.0 | 5.8 | $0.184 | 1/1 | 4.46s |
| #145 | GPT-5.4 none | OpenAI | 10.0 | 5.8 | $0.397 | 1/1 | 2.75s |
| #146 | Nemotron 3 Super medium | NVIDIA | 10.0 | 5.7 | $0.055 | 1/1 | 39.7s |
| #147 | GLM 5 none | Z.ai | 10.0 | 5.7 | $0.041 | 1/1 | 11.1s |
| #148 | Qwen3.5-122B-A10B none | Qwen | 10.0 | 5.7 | $0.247 | 1/1 | 2.04s |
| #149 | Gemini 3.1 Flash Lite high | 10.0 | 5.6 | $2.044 | 1/1 | 6.44s | |
| #150 | KAT-Coder-Air V2.5 high | Kwaipilot | 10.0 | 5.6 | $0.077 | 1/1 | 4.77s |
| #151 | GLM 5V Turbo none | Z.ai | 10.0 | 5.6 | $0.052 | 1/1 | 4.86s |
| #152 | Owl Alpha medium | Openrouter | 10.0 | 5.6 | $0.000 | 1/1 | 8.26s |