Technical Tutorials

How to Deploy and Self-Host Llama 3 on a Cheap Cloud VPS with Ollama and Docker

Step-by-step technical guide to self-hosting quantized Llama 3 8B on an affordable VPS instance with Ollama, Open WebUI, and secure SSL.

Equipa de Sistemas @ NexaLabs

2026-09-0410 min read

How to Deploy and Self-Host Llama 3 on a Cheap Cloud VPS with Ollama and Docker

1. Hardware Prerequisites & VPS Selection

To run quantized Llama 3 8B (Q4_K_M) at 20-30 tokens/sec on CPU inference, you need a minimum of 4 modern vCPUs (AMD EPYC or Ryzen) and 8 GB RAM with NVMe storage.

Verified Partner Offer

AlphaVPS

High-frequency AMD Ryzen nodes with 10% lifetime discount

Check Ryzen VPS Plans on AlphaVPS (from €3.50/mo)

Affiliate link: we earn a commission if you subscribe, at no extra cost to you.

2. Docker Compose Configuration for Ollama & Open WebUI

Create a project directory on your Linux instance and launch this compose stack:

docker-compose.yml

version: '3.8'

services:
  ollama:
    image: ollama/ollama:latest
    container_name: nexalabs-ollama
    restart: unless-stopped
    ports:
      - "11434:11434"
    volumes:
      - ./ollama_data:/root/.ollama
    environment:
      - OLLAMA_KEEP_ALIVE=24h
      - OLLAMA_NUM_PARALLEL=4

  open-webui:
    image: ghcr.io/open-webui/open-webui:main
    container_name: nexalabs-webui
    restart: unless-stopped
    ports:
      - "3000:8080"
    environment:
      - OLLAMA_BASE_URL=http://ollama:11434
    volumes:
      - ./webui_data:/app/backend/data
    depends_on:
      - ollama

3. Model Download and REST API Testing

Run the following commands to pull and verify your self-hosted LLM endpoint:

Terminal do Servidor

# Levantar los contenedores en segundo plano
docker compose up -d

# Descargar Llama 3 (8B) optimizado
docker exec -it nexalabs-ollama ollama run llama3

# Probar la API compatible con OpenAI desde curl
curl http://localhost:11434/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama3",
    "messages": [{"role": "user", "content": "Explica la computación cuántica en una frase"}]
  }'

Frequently Asked Questions

How much does it cost to host this monthly?

On AlphaVPS or Hostinger, an 8GB RAM Ryzen node costs around €7–€12/month, with zero per-token API charges.