Tabby

A self-hosted, open-source AI coding assistant that supports multiple LLM backends and runs entirely on-premise for maximum security.

Overview

Tabby allows organizations to deploy their own AI coding assistant using open-source models like StarCoder or Llama. It provides autocomplete and chat features within VS Code and JetBrains, ensuring that no code data leaves the internal network.

Best for

• Security-conscious enterprises
• Teams with existing GPU infrastructure
• Cost-sensitive deployments with local LLMs

Trade-offs

What XeroHack pre-wires

When you pick Tabby in the interview, the engine emits these files into your scaffold:

• Default local model selection
• VS Code extension config
• JetBrains plugin config