Show HN: Janus – Go binary that runs GGUF models via Vulkan on AMD/Intel/Nvidia

Janus is a Go-based application that runs GGUF language models locally on Windows, Linux, and macOS using Vulkan acceleration (or CPU fallback), exposing an OpenAI-compatible API and web interface without requiring Python, Docker, or Ollama. It includes built-in tools for file operations, command execution, PDF/document handling, and OCR, while allowing users to swap models and access the service through command-line clients, IDE integrations, or a web UI. The project is designed for local inference with optional authentication and requires Go 1.22+, a Vulkan-capable GPU (recommended), and disk space for model files.