vLLM Launcher — Windows Desktop Workbench for Local LLMs (vLLM, SGLang, llama.cpp)

VLLM Launcher, a Windows desktop app for managing local LLM inference through WSL2, has been released as v0.1.0-alpha. The app supports multiple engines including vLLM, SGLang, and llama.cpp, and provides an OpenAI-compatible and Anthropic-compatible API gateway. It is available as a portable Windows release (~31 MB) and is built with Go/Wails, React/TypeScript, and Node.js.

Hi everyone, I’m sharing a project I’ve been building: vLLM Launcher — a Windows desktop app for managing local LLM inference through WSL2. What it does: Multi-engine management: vLLM, SGLang, llama.cpp from one desktop UI Local API gateway with OpenAI-compatible and Anthropic-compatible endpoints Instance cards for start/stop/restart/inspect workflows Request logs, health checks, runtime metrics, and disk model discovery Floating mode for compact status while working in other apps WSL process orchestration that survives Node.js restarts Tech stack: Go / Wails desktop shell React + TypeScript frontend Node.js API service Model engines running inside WSL2 Download: Windows portable release ~31 MB, unzip and run This is the first public release v0.1.0-alpha . Feedback and setup issues are especially welcome since I’m still refining the installation flow across different Windows/WSL/driver setups. Would love to hear from anyone running local models on Windows