PD Serving of Qwen3.8-2.4T
How vLLM reaches 5K throughput and 180 interactivity on Qwen3.8-2.4T with GB300 NVL72 PD serving and how to reproduce results yourself.
阅读 vLLM 官方博客（网页） 原文 ↗How vLLM reaches 5K throughput and 180 interactivity on Qwen3.8-2.4T with GB300 NVL72 PD serving and how to reproduce results yourself.
阅读 vLLM 官方博客（网页） 原文 ↗