SGLang – Fast, Open-Source LLM & Multimodal Serving Framework

SGLang powers fast, scalable inference for large language and multimodal models. Open-source serving framework with state-of-the-art performance.

关键词：SGLang, LLM inference, LLM serving, serving framework, large language models, multimodal models, vision language models, RadixAttention, GPU inference, open source, AI infrastructure, model deployment