Skip to main content

​
GAIA Lightweight Installer - Implementation Plan

Status: Planning Priority: High View full plan on GitHub • Vote with 👍

​
Executive Summary

Create a lightweight, cross-platform installer for GAIA core components that’s as easy as installing uv: 
irm https://amd-gaia.ai/install.ps1 | iex
Goal: Zero to chatting in under 2 minutes.

​
What’s Included

​
Core (Included by Default)

ComponentCommandDescription
CLI BasegaiaCore framework and commands
Chatgaia chatChat SDK with RAG
Code Agentgaia-codeAI coding assistant
MCPgaia mcpModel Context Protocol servers

​
Extended (Install Separately)

gaia install emr      # Medical intake agent
gaia install talk     # Voice agent (ASR/TTS)
gaia install blender  # Blender automation
gaia install jira     # Jira integration
gaia install eval     # Evaluation framework

​
Installation Channels

One-Liner

PowerShell/bash script Zero dependencies ~30 second install

winget

Native Windows package Auto-updates winget install AMD.GAIA

pip/uv

For developers Integrates with venvs uv pip install amd-gaia

​
Self-Management Commands

​
Update GAIA

# Update to latest
gaia update

# Check for updates without installing
gaia update --check

# Update to specific version
gaia update --version 0.16.0

# Use beta channel
gaia update --channel beta

​
System Diagnostics

# Run health checks
gaia doctor
Output: 
GAIA System Diagnostics
========================================

  ✓ Python version: 3.11.0
  ✓ GAIA installation: 0.16.0
  ✓ Lemonade server: Running
  ✓ Default model: Loaded
  ✓ Disk space: 45 GB available
  ✓ Memory: 12 GB available
  ✓ Network: Connected

All checks passed!

​
Uninstall

# Remove GAIA (keep data)
gaia uninstall

# Remove everything including data
gaia uninstall --purge

​
Installation Flow

​
Package Structure

[project.optional-dependencies]
# Core - minimal
core = ["click", "rich", "httpx", "pydantic"]

# Chat - adds RAG
chat = ["pypdf", "sentence-transformers", "faiss-cpu"]

# MCP integration
mcp = ["websockets"]

# Default install
default = ["amd-gaia[core]", "amd-gaia[chat]", "amd-gaia[mcp]"]

# Extended (install separately)
emr = ["amd-gaia[chat]", "pillow"]
talk = ["sounddevice", "numpy"]
eval = ["amd-gaia[chat]", "pandas"]

​
Platform Support

​
Windows

  • PowerShell one-liner
  • winget package
  • Start Menu shortcuts
  • System tray integration (optional)

​
Linux

  • Bash one-liner
  • .desktop file for app launchers
  • Shell completion (bash, zsh, fish)

​
macOS

  • Bash one-liner
  • Homebrew formula (future)

​
Success Metrics

MetricTarget
Install time (one-liner)< 60 seconds
Update time< 30 seconds
Install script size< 10 KB
Core package size< 50 MB
Support tickets about install< 5% of users

Full Implementation Plan

View the complete technical specification on GitHub