Skip to content

Welcome to Backend.AI GO

Backend.AI GO is a user-friendly desktop application designed to let you run Large Language Models (LLMs) like Gemma 4, Qwen3.6, gpt-oss-120B, Qwen3.5-122B-A10B, Solar-Open-100B, and Mistral directly on your local machine.

For developers looking for a local AI sandbox, privacy-conscious professionals, and AI enthusiasts, Backend.AI GO provides the tools to download, manage, and interact with the latest AI models without relying on cloud services.

Key Highlights

  • Privacy First: Your data stays on your machine. No cloud uploads required for local inference.
  • Hardware Optimized: High performance on macOS (Metal/MLX), Windows/Linux (NVIDIA CUDA, AMD ROCm), and standard CPUs.
  • Agentic Capabilities: Use built-in agents that can perform tasks using tools like web search, file operations, and more.
  • Easy Management: Search and download models directly from Hugging Face within the app.
  • Open Standard API: Provides an OpenAI-compatible API so you can use Backend.AI GO as a backend for your favorite AI tools.

How to use this documentation

If you are new to Backend.AI GO, we recommend starting with the Getting Started section.

  • Quickstart: Get up and running in less than 5 minutes.
  • Cowork: Learn how to use autonomous agents.
  • CLI Reference: For those who prefer the terminal.

Developed and maintained as part of the Backend.AI project.