通过单一API网关访问和管理多个大型语言模型,提供统一仪表板、成本控制和智能路由功能。
| 模型 | 速度 | 延迟 | 测试数 |
|---|---|---|---|
| qwen/qwen3-next-80b-a3b-instruct | 177.68 t/s | 1.31s | 5 |
| qwen/qwen3-next-80b-a3b-instruct | 177.68 t/s | 1.31s | 5 |
| deepseek/deepseek-v3 | 0.00 t/s | 0.00s | 5 |
| google/gemini-2.5-pro | 0.00 t/s | 0.00s | 5 |
| meta-llama/llama-4-maverick | 0.00 t/s | 0.00s | 5 |
| meta-llama/llama-4-scout | 0.00 t/s | 0.00s | 5 |
| mistralai/ministral-3b | 0.00 t/s | 0.00s | 5 |
| mistralai/mistral-large | 0.00 t/s | 0.00s | 5 |
| 时间 | 模型 | 速度 | 延迟 |
|---|---|---|---|
| Oct 31, 11:22 AM | mistralai/mistral-large | 0.00 t/s | 0.00s |
| Oct 31, 11:22 AM | meta-llama/llama-4-maverick | 0.00 t/s | 0.00s |
| Oct 31, 11:21 AM | meta-llama/llama-4-scout | 0.00 t/s | 0.00s |
| Oct 31, 10:49 AM | deepseek/deepseek-v3 | 0.00 t/s | 0.00s |
| Oct 31, 10:48 AM | mistralai/ministral-3b | 0.00 t/s | 0.00s |
| Oct 31, 10:47 AM | google/gemini-2.5-pro | 0.00 t/s | 0.00s |
| Oct 27, 02:09 AM | qwen/qwen3-next-80b-a3b-instruct | 177.68 t/s | 1.31s |