SGLang

May 10, 2026 · View on GitHub

SGLang is a high-performance serving framework for large language models (LLMs) and vision-language models (VLMs). By co-designing the backend runtime and a frontend language, it delivers faster and more controllable interactions. See the SGLang repository and documentation for details.

This directory contains SGLang deployment guides for different MiniCPM versions. Use the links below to jump to the specific version.

Versioned Deployment Guides

VersionEnglish中文
MiniCPM-V 4.6 (latest)Guide指南
MiniCPM-V 4.5Guide指南
MiniCPM-V 4.0Guide指南
MiniCPM-o 4.5Guide指南