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 ↓.
210/210
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） |
|---|---|---|---|---|---|---|---|
| #35 | Seed-2.0-Lite medium | Bytedance Seed | 10.0 | 7.9 | $0.234 | 1/1 | 12.4s |
| #36 | Qwen3.7 Plus medium | Qwen | 10.0 | 7.9 | $0.267 | 1/1 | 15.0s |
| #37 | Qwen3.6 Plus medium | Qwen | 10.0 | 7.8 | $0.405 | 1/1 | 5.87s |
| #38 | GLM 5.2 medium | Z.ai | 10.0 | 7.8 | $0.222 | 1/1 | 20.4s |
| #39 | GPT-5.6 Terra medium | OpenAI | 10.0 | 7.8 | $0.676 | 1/1 | 5.09s |
| #40 | Claude Sonnet 4.6 medium | Anthropic | 10.0 | 7.8 | $2.057 | 1/1 | 7.48s |
| #41 | Claude Opus 4.8 low | Anthropic | 10.0 | 7.8 | $2.077 | 1/1 | 6.85s |
| #42 | GLM 5 medium | Z.ai | 10.0 | 7.7 | $0.307 | 1/1 | 15.9s |
| #43 | Claude Opus 4.6 medium | Anthropic | 10.0 | 7.7 | $3.059 | 1/1 | 9.73s |
| #44 | GPT-5.6 Luna high | OpenAI | 10.0 | 7.7 | $1.017 | 1/1 | 4.98s |
| #45 | DeepSeek V4 Flash high | DeepSeek | 10.0 | 7.7 | $0.042 | 1/1 | 74.7s |
| #46 | DeepSeek V4 Pro high | DeepSeek | 9.8 | 7.7 | $0.200 | 1/1 | 15.9s |
| #47 | MiniMax M3 medium | Minimax | 10.0 | 7.6 | $0.286 | 1/1 | 11.9s |
| #48 | Grok Build 0.1 medium | X AI | 10.0 | 7.6 | $1.097 | 1/1 | 13.1s |
| #49 | GLM 5 Turbo medium | Z.ai | 10.0 | 7.6 | $0.323 | 1/1 | 9.84s |