🧠 Local AI Packaged

⚠️ Note: This project is a maintained and extended fork of coleam00/local-ai-packaged, which no longer appears to be actively maintained.
This version includes up‑to‑date dependencies, bug fixes, and full automation scripts for setup and management.

---

🌍 Overview

Local AI Packaged provides a full self‑hosted AI environment using Docker Compose.
It bundles everything you need for local LLM workflows, including:

• 🧠 Ollama for local LLM inference
• 💬 Open WebUI for chatting with your models or n8n agents
• ⚙️ n8n for building automation workflows
• 🧱 Supabase as database, vector store, and auth system
• 🧩 Flowise, Langfuse, Qdrant, Neo4j, SearXNG, and optional Caddy

This version is designed for technical self‑hosters running the stack on a home server, NAS (e.g. Synology), or any Linux host.

---

⚙️ Prerequisites

Make sure you have:

• 🐳 Docker and Docker Compose installed
• 🐍 Python 3.8+
• 💾 Git
• 💡 At least 16 GB RAM recommended

GPU (optional)

• NVIDIA: Install NVIDIA Container Toolkit
• AMD: ROCm runtime configured
• CPU only: Works fine, just slower

---

🚀 Quick Installation

① Clone the repository

git clone https://github.com/gamersalpha/local-ai-packaged.git
cd local-ai-packaged

② Generate the .env automatically

The generate_env.py script creates a full environment configuration automatically:

• Generates secure random secrets
• Detects GPU type (NVIDIA / AMD / CPU)
• Sets correct local paths for volumes

Run:

python3 generate_env.py --yes --regen-sensitive --docker

Once complete, it will start the stack and show the service URLs:

🌐 Available services:
  🧠 Ollama      : http://192.168.1.42:11434
  ⚙️ n8n          : http://192.168.1.42:5678
  💬 Open WebUI  : http://192.168.1.42:8080
  🧱 Supabase    : http://192.168.1.42:54323

---

🧩 Managing the stack

▶️ Start automatically

Use the provided Python launcher for easy orchestration:

python3 start_services.py [options]

Options

Option	Description
`--profile [cpu	gpu-nvidia
`--environment [private	public]`
--no-supabase	Skip Supabase include
--no-caddy	Skip Caddy reverse proxy
--update	Pull latest Docker images before start
--dry-run	Preview configuration only

Examples

# CPU-only deployment
python3 start_services.py --profile cpu

# NVIDIA GPU
python3 start_services.py --profile gpu-nvidia

# Skip Supabase and Caddy
python3 start_services.py --no-supabase --no-caddy

# Pull new images before starting
python3 start_services.py --update

💡 What it does:

• Validates or creates your .env
• Clones Supabase’s official Docker stack if missing
• (De)comments Caddy and Supabase dynamically
• Generates a new secret key for SearXNG
• Stops existing containers before redeploy
• Starts Supabase → then the Local AI stack

---

♻️ Update all services

Use the helper script update_services.sh:

./update_services.sh

This will:

1. Stop all containers
2. Pull the latest images
3. Restart the stack via start_services.py

For GPU users, you can adjust:

docker compose -p localai -f docker-compose.yml --profile gpu-nvidia down
docker compose -p localai -f docker-compose.yml --profile gpu-nvidia pull
python3 start_services.py --profile gpu-nvidia --no-caddy

---

🌐 Access your local services

Service	Description	Default URL
n8n	Workflow automation	http://192.168.1.42:5678
Open WebUI	Chat interface for LLMs	http://192.168.1.42:8080
Ollama	Local LLM API	http://192.168.1.42:11434
Flowise	Low‑code AI builder	http://192.168.1.42:3001
Supabase	DB, Auth & Vector store	http://192.168.1.42:54323
Langfuse	LLM tracing dashboard	http://192.168.1.42:3000
SearXNG	Web search engine for RAG	http://192.168.1.42:8080
Neo4j	Graph database	http://192.168.1.42:7474

(Adjust IPs for your own LAN setup.)

---

🛠️ Troubleshooting

Supabase fails to start

• Check that the folder supabase/ was created automatically.
• Delete it and rerun:

  python3 start_services.py --no-caddy

• Ensure .env contains:

  POOLER_DB_POOL_SIZE=5

GPU not detected

• Ensure the NVIDIA Container Toolkit or ROCm is installed correctly.
• Fallback to CPU mode if needed:

  python3 start_services.py --profile cpu

Ports already in use

Edit docker-compose.yml to change exposed ports, then restart.

---

🧾 Project structure

.
├── docker-compose.yml
├── start_services.py
├── update_services.sh
├── generate_env.py
├── supabase/           # auto-cloned (ignored by Git)
├── n8n/
│   └── backup/
├── searxng/
├── shared/
└── neo4j/

📝 Note: The supabase/ folder is automatically cloned by start_services.py from github.com/supabase/supabase.
It should not be committed to Git and is already included in .gitignore.

---

📜 License

Licensed under the Apache 2.0 License.
See LICENSE for details.

---

Built and maintained with ❤️ for the self‑hosting community.