Hugging Face launches Llama for Windows to run AI models locally

Hugging Face and the llama.cpp development team have released ‘Llama for Windows’, a new application designed to allow Windows 11 users to download and run large language models (LLMs) locally on their computers. The tool aims to reduce reliance on cloud-based services like ChatGPT or Gemini, offering an alternative for users prioritizing privacy or seeking to avoid subscription fees.

Unlike previous methods that required complex command-line operations, Llama for Windows provides a simplified interface via the Windows taskbar tray. Built using Microsoft’s WinUI 3 framework, the app allows users to browse and download suggested models from the Hugging Face Hub with a single click. Once a model is loaded, users can interact with it through a web-based chat interface.

The application also includes optimizations for NVIDIA RTX Spark-equipped PCs and runs natively on Windows on Arm. For developers, the tool functions as a local server with an OpenAI-compatible API, enabling custom applications, coding editors, and AI agents to utilize locally installed models without sending data to external cloud providers.