为MBZUAI-IFM/K2-Think模型提供API代理服务,具备令牌管理和负载均衡功能。
| 模型 | 速度 | 延迟 | 测试数 |
|---|---|---|---|
| MBZUAI-IFM/K2-Think | 229.66 t/s | 2.22s | 15 |
| MBZUAI-IFM/K2-Think-nothink | 220.82 t/s | 2.18s | 10 |
| 时间 | 模型 | 速度 | 延迟 |
|---|---|---|---|
| Nov 3, 12:16 PM | MBZUAI-IFM/K2-Think-nothink | 219.47 t/s | 2.56s |
| Oct 30, 06:56 PM | MBZUAI-IFM/K2-Think | 227.48 t/s | 2.43s |
| Oct 17, 05:41 PM | MBZUAI-IFM/K2-Think-nothink | 222.17 t/s | 1.80s |
| Sep 12, 09:46 AM | MBZUAI-IFM/K2-Think | 224.98 t/s | 2.05s |
| Sep 12, 09:10 AM | MBZUAI-IFM/K2-Think | 236.50 t/s | 2.18s |