LMSYS:Blog(Chatbot Arena 团队)·· 13 小时前AI 评分40
在 GB300 NVL72 上部署 DeepSeek:长上下文推理大幅提升
Blog Deploying DeepSeek on GB300 NVL72: Big Wins in Long-Context Inference As the latest addition to the Blackwell family, the GB300 NVL72 is the most powerful platform for long-context LLM inference. In this blog post, we share our latest progress on optimizing DeepSeek R1-... Nvidia & SGLang Team February 19, 2026
AI 导读
NVIDIA 与 SGLang 团队公布在 GB300 NVL72 上优化 DeepSeek R1-NVFP4 长上下文推理的进展,128K/8K 场景下 SGLang 达到 226.2 TPS/GPU,较 GB200 提升 1.53 倍。
来源:LMSYS:Blog(Chatbot Arena 团队) · lmsys.org