Cloud vs Self-Hosted

Vega AI offers two deployment options: a fully managed cloud service and self-hosted installation. Choose the option that best fits your privacy requirements, technical expertise, and usage patterns.

Quick Comparison

Feature	Cloud	Self-Hosted
Setup Time	Instant	5 minutes
Maintenance	Zero	Minimal
Data Privacy	Managed by service	Complete control
AI Quota	10 analyses/month	Unlimited
Authentication	Google OAuth	Local accounts
Updates	Automatic	Manual pull
Cost	Free tier	API costs only
Customization	Limited	Full control

Cloud Deployment

The cloud version is hosted at vega.benidevo.com and provides instant access without any setup.

Advantages

Zero Setup

Start tracking jobs immediately with no installation required

Always Updated

Automatic updates with the latest features and fixes

Cross-Device Access

Access your data from anywhere with internet

No Maintenance

No need to manage infrastructure or updates

Features

Authentication

Google OAuth only (no password-based accounts)
Secure session management with JWT tokens
Automatic user creation on first login

AI Usage Quotas

10 AI job analyses per month (free tier)
Unlimited job tracking and management
Admin users have unlimited quotas

Data Isolation

Multi-tenant architecture with row-level security
Your data is isolated from other users
Shared company reference data (prevents duplicates)

Limitations

Quota Limits: Free tier includes 10 AI analyses per month. Self-hosted has no limits.

Data Privacy: While your data is isolated and secure, it’s stored on our servers. For complete data control, use self-hosted mode.

Self-Hosted Deployment

Run Vega AI on your own infrastructure with complete control over your data and configuration.

Advantages

Complete Privacy

Your data never leaves your infrastructure

Unlimited AI Usage

No quotas on AI features (only Gemini API limits apply)

Full Customization

Configure every aspect to your needs

No Internet Required

Run locally without external dependencies (except Gemini API)

Requirements

Minimum Setup:

Docker (or compatible runtime)
Gemini API key (free from Google)
100MB disk space
512MB RAM

Optional:

Docker Compose (for easier management)
Reverse proxy (for HTTPS/custom domain)
External database (default: SQLite)

Features

Authentication

Local username/password accounts
Default admin account created on first run
Optional Google OAuth integration

Configuration

config

# Minimal configuration
GEMINI_API_KEY=your-api-key

# Optional: customize admin credentials
ADMIN_USERNAME=myadmin
ADMIN_PASSWORD=SecurePassword123

# Optional: advanced settings
TOKEN_SECRET=your-jwt-secret
CORS_ALLOWED_ORIGINS=*

Data Storage

SQLite database with optimized settings
Docker volume for persistence
Easy backup and restore
Data portability

Deployment Options

Docker Run
Docker Compose
Docker Swarm

docker run --pull always -d \
  --name vega-ai \
  -p 8765:8765 \
  -v vega-data:/app/data \
  -e GEMINI_API_KEY=your-api-key \
  ghcr.io/benidevo/vega-ai:latest

docker-compose.yml

services:
  vega-ai:
    image: ghcr.io/benidevo/vega-ai:latest
    ports:
      - "8765:8765"
    volumes:
      - vega-data:/app/data
    environment:
      - GEMINI_API_KEY=your-api-key
    restart: unless-stopped

volumes:
  vega-data:

Start with: docker compose up -d

docker-compose.yml

services:
  vega-ai:
    image: ghcr.io/benidevo/vega-ai:latest
    ports:
      - "8765:8765"
    volumes:
      - vega-data:/app/data
    secrets:
      - gemini_api_key
    environment:
      - GEMINI_API_KEY_FILE=/run/secrets/gemini_api_key
    deploy:
      replicas: 1
      restart_policy:
        condition: on-failure

secrets:
  gemini_api_key:
    external: true

volumes:
  vega-data:

Deploy with secrets:

echo "your-api-key" | docker secret create gemini_api_key -
docker stack deploy -c docker-compose.yml vega-stack

Platform Support

Vega AI Docker images support multiple architectures:

AMD64 (Intel/AMD processors)
ARM64 (Apple Silicon, Raspberry Pi 4+)

The same image works across all platforms:

# Works on Intel, AMD, and ARM processors
docker pull ghcr.io/benidevo/vega-ai:latest

Cost Comparison

Cloud Mode

Free Tier:

10 AI analyses per month
Unlimited job tracking
All features included

Your Costs:

$0 (free tier)
No infrastructure costs
No API costs

Self-Hosted Mode

Infrastructure:

Free (run on existing hardware)
Or $5-10/month (small VPS)

API Costs:

Gemini API usage (pay-as-you-go)
Typically $0-5/month for personal use
First 15 requests/minute are free

Gemini API Pricing: Google’s Gemini API has a generous free tier. For typical job search usage (5-10 analyses per week), costs are usually zero or minimal. Check current pricing at Google AI Pricing.

Migration Between Modes

Cloud to Self-Hosted

Currently, you’ll need to manually recreate your data:

Export job data from cloud (copy/paste)
Download your profile information
Set up self-hosted instance
Import data manually

Future Feature: Automated data export/import is planned for a future release.

Self-Hosted to Cloud

Same process as above - manual migration currently required.

Security Considerations

Cloud Mode

Pros:

Managed infrastructure and security updates
HTTPS/TLS by default
Professional monitoring and backups

Cons:

Data stored on external servers
Subject to service availability
Must trust service provider

Self-Hosted Mode

Pros:

Complete data control
No third-party access
Air-gapped deployment possible

Cons:

You manage security updates
Need to configure HTTPS yourself
Responsible for backups

Recommendations

Choose Cloud If:

✅ You want instant setup with zero maintenance ✅ You’re comfortable with managed services ✅ You don’t need more than 10 AI analyses per month ✅ You prefer automatic updates ✅ You want cross-device access without VPN

Choose Self-Hosted If:

✅ You need complete data privacy ✅ You want unlimited AI usage ✅ You have technical expertise (or want to learn) ✅ You need custom configurations ✅ You’re already running other self-hosted services ✅ You want to run it air-gapped or on a local network

Getting Started

Try Cloud

Start using Vega AI instantly with no setup

Self-Host Guide

Deploy Vega AI on your own infrastructure

Can’t Decide? Start with the cloud version to test Vega AI. You can always migrate to self-hosted later if your needs change.

Overview

Getting Started

Core Features

Deployment

Integrations

​Quick Comparison

​Cloud Deployment

​Advantages

Zero Setup

Always Updated

Cross-Device Access

No Maintenance

​Features

​Limitations

​Self-Hosted Deployment

​Advantages

Complete Privacy

Unlimited AI Usage

Full Customization

No Internet Required

​Requirements

​Features

​Deployment Options

​Platform Support

​Cost Comparison

​Cloud Mode

​Self-Hosted Mode

​Migration Between Modes

​Cloud to Self-Hosted

​Self-Hosted to Cloud

​Security Considerations

​Cloud Mode

​Self-Hosted Mode

​Recommendations

​Choose Cloud If:

​Choose Self-Hosted If:

​Getting Started

Try Cloud

Self-Host Guide

Build docs developers (and LLMs) love

Quick Comparison

Cloud Deployment

Advantages

Features

Limitations

Self-Hosted Deployment

Advantages

Requirements

Features

Deployment Options

Platform Support

Cost Comparison

Cloud Mode

Self-Hosted Mode

Migration Between Modes

Cloud to Self-Hosted

Self-Hosted to Cloud

Security Considerations

Cloud Mode

Self-Hosted Mode

Recommendations

Choose Cloud If:

Choose Self-Hosted If:

Getting Started