{"id":306,"date":"2026-11-05T17:34:38","date_gmt":"2026-11-05T17:34:38","guid":{"rendered":"https:\/\/giganodes.host\/blog\/?p=306"},"modified":"2026-11-05T17:34:38","modified_gmt":"2026-11-05T17:34:38","slug":"host-n8n-and-ollama-on-one-vps-india","status":"publish","type":"post","link":"https:\/\/giganodes.host\/blog\/host-n8n-and-ollama-on-one-vps-india","title":{"rendered":"How to Host n8n and Ollama on One VPS in India (Step by Step, 2027)"},"content":{"rendered":"

Last updated: 5 November 2026. Plan prices are from giganodes.host\/vps\/india and giganodes.host\/vps\/india\/n8n-vps on that date and exclude 18% GST. Software versions (Ubuntu 24.04, n8n, Ollama, Caddy) were current when written; re-check the images before you deploy.<\/em><\/p>\n

n8n gives you the automation glue, Ollama gives you a language model that never leaves your server, and a single VPS in India keeps both inside the country at a fixed rupee price. This tutorial walks through the whole build: picking a plan that can honestly hold a quantised 3B, 7B or 8B model next to n8n, installing Docker on Ubuntu 24.04, a Compose file that puts n8n, Ollama, PostgreSQL and Caddy behind HTTPS, wiring the Ollama node, a Telegram-to-LLM-to-Google-Sheets workflow, then backups, firewalling and updates. Nothing here needs a GPU.<\/p>\n

\n

TL;DR:<\/strong> Use Cloud VPS S<\/strong> (4 vCore EPYC 7C13, 8 GB RAM, \u20b91,800\/mo or \u20b91,584\/mo annual) for n8n plus a 3B quantised model, or a 7B\/8B Q4 model with little headroom. Use Cloud VPS M<\/strong> (6 vCore, 16 GB, \u20b93,600\/mo or \u20b93,168\/mo annual) if you want a 7B\/8B model to run comfortably alongside PostgreSQL and busy workflows. CPU-only inference is slow: expect single-digit tokens per second. Put Ollama on the internal Docker network only, expose n8n through Caddy on 443, lock everything else with UFW, and back up the PostgreSQL dump plus the n8n data volume nightly.<\/p>\n<\/div>\n

Which VPS Plan Fits n8n Plus Ollama in India?<\/h2>\n

The honest answer is 8 GB of RAM for a 3B model, 16 GB if you want a 7B or 8B model to run without squeezing n8n and PostgreSQL. Ollama loads the whole quantised model into RAM; a 3B model at 4-bit quantisation is roughly 2 GB on disk and needs a little more than that in memory once context is added, while 7B and 8B models at Q4 are roughly 4.5 to 5 GB on disk and want 6 GB or more in use. n8n itself, PostgreSQL, Caddy and Ubuntu take about 1.5 to 2 GB between them.<\/p>\n

These are the AMD EPYC 7C13 plans on giganodes.host\/vps\/india<\/a>, which are the same plans listed on the n8n VPS page<\/a>. Every plan has NVMe storage, a dedicated IPv4, full root access and Cloudflare Magic Transit DDoS protection included. The n8n page starts its recommendations at Cloud VPS S, as does our n8n VPS India guide<\/a>, and after testing the stack we agree: Cloud Nano (1 vCore, 2 GB) and Cloud XS (2 vCore, 4 GB) can run n8n alone, but not n8n and a useful model together.<\/p>\n

\n\n\n\n\n\n\n\n\n\n
Plan<\/th>\nvCores \/ RAM \/ NVMe<\/th>\nMonthly<\/th>\nAnnual (per month)<\/th>\nModel size that fits with n8n<\/th>\nOur verdict<\/th>\n<\/tr>\n<\/thead>\n
Cloud XS<\/td>\n2 \/ 4 GB \/ 40 GB<\/td>\n\u20b91,000\/mo<\/td>\n\u20b9880\/mo<\/td>\n1B class only; not recommended for this stack<\/td>\nn8n only, no Ollama<\/td>\n<\/tr>\n
Cloud VPS S<\/td>\n4 \/ 8 GB \/ 60 GB<\/td>\n\u20b91,800\/mo<\/td>\n\u20b91,584\/mo<\/td>\n3B Q4 comfortably; 7B\/8B Q4 with little headroom<\/td>\nMinimum we recommend<\/td>\n<\/tr>\n
Cloud VPS M<\/td>\n6 \/ 16 GB \/ 120 GB<\/td>\n\u20b93,600\/mo<\/td>\n\u20b93,168\/mo<\/td>\n7B\/8B Q4 comfortably; room for two loaded models<\/td>\nBest fit for most teams<\/td>\n<\/tr>\n
Cloud VPS L<\/td>\n12 \/ 32 GB \/ 256 GB<\/td>\n\u20b97,200\/mo<\/td>\n\u20b96,336\/mo<\/td>\n13B\/14B Q4; several models loaded at once<\/td>\nOnly if you need larger models<\/td>\n<\/tr>\n
Cloud VPS XL<\/td>\n20 \/ 64 GB \/ 512 GB<\/td>\n\u20b914,400\/mo<\/td>\n\u20b912,672\/mo<\/td>\n30B class Q4 (still CPU-slow)<\/td>\nCompare with a dedicated server first<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

Prices exclude 18% GST; a GST invoice is issued on every payment. Annual billing saves 12%.<\/small><\/p>\n

Two cautions before you buy. First, CPU-only inference is slow. On any of these plans, expect single-digit tokens per second from a 7B\/8B model, faster from a 3B. That is fine for a Telegram bot, a summariser or a classifier that runs in the background; it is not fine for a chat UI where a human is waiting. Second, more vCores help throughput but do not change that picture much; RAM capacity is what decides which model you can load at all. If you need more than a few tokens per second, read our Ollama on VPS guide<\/a> for the trade-offs, and our VPS sizing guide<\/a> for the general method.<\/p>\n

Step 1: Prepare Ubuntu 24.04 and Install Docker<\/h2>\n

Start from a fresh Ubuntu 24.04 LTS image, create a non-root user, and install Docker from Docker’s own repository so Compose v2 comes with it. Pick Ubuntu 24.04 when ordering in the VirtFusion panel; the VPS is live within about a minute.<\/p>\n

ssh root@YOUR_SERVER_IP\napt update && apt upgrade -y\nadduser deploy\nusermod -aG sudo deploy\nrsync --archive --chown=deploy:deploy ~\/.ssh \/home\/deploy\n<\/code><\/pre>\n
Log back in as deploy<\/code> and install Docker Engine and the Compose plugin.<\/p>\n
sudo apt install -y ca-certificates curl gnupg\nsudo install -m 0755 -d \/etc\/apt\/keyrings\ncurl -fsSL https:\/\/download.docker.com\/linux\/ubuntu\/gpg | sudo gpg --dearmor -o \/etc\/apt\/keyrings\/docker.gpg\necho \"deb [arch=$(dpkg --print-architecture) signed-by=\/etc\/apt\/keyrings\/docker.gpg] https:\/\/download.docker.com\/linux\/ubuntu $(. \/etc\/os-release && echo $VERSION_CODENAME) stable\" | sudo tee \/etc\/apt\/sources.list.d\/docker.list > \/dev\/null\nsudo apt update\nsudo apt install -y docker-ce docker-ce-cli containerd.io docker-compose-plugin\nsudo usermod -aG docker deploy\nnewgrp docker\ndocker compose version\n<\/code><\/pre>\n
Point a DNS A record at the server now, for example n8n.yourdomain.in<\/code>, so Caddy can issue a certificate the moment it starts. If you proxy through Cloudflare, leave the record on “DNS only” (grey cloud) until the first certificate is issued.<\/p>\n
Step 2: Create the Project and Environment File<\/h2>\n
Keep everything in one directory with a .env<\/code> file that Compose reads automatically. The encryption key is what n8n uses to encrypt stored credentials; lose it and every saved API key in n8n becomes unreadable, so copy it into your password manager.<\/p>\n
mkdir -p ~\/n8n-ollama && cd ~\/n8n-ollama\nopenssl rand -hex 32   # use this for N8N_ENCRYPTION_KEY\nopenssl rand -hex 16   # use this for POSTGRES_PASSWORD\nnano .env\n<\/code><\/pre>\n
# .env\nDOMAIN=n8n.yourdomain.in\nACME_EMAIL=you@yourdomain.in\n\nN8N_ENCRYPTION_KEY=paste-64-hex-chars-here\nGENERIC_TIMEZONE=Asia\/Kolkata\nTZ=Asia\/Kolkata\n\nPOSTGRES_USER=n8n\nPOSTGRES_PASSWORD=paste-32-hex-chars-here\nPOSTGRES_DB=n8n\n\n# Ollama tuning for a CPU-only box\nOLLAMA_KEEP_ALIVE=30m\nOLLAMA_NUM_PARALLEL=1\nOLLAMA_MAX_LOADED_MODELS=1\n<\/code><\/pre>\n
OLLAMA_MAX_LOADED_MODELS=1<\/code> and OLLAMA_NUM_PARALLEL=1<\/code> matter on an 8 GB plan: they stop Ollama from trying to hold two models or two concurrent generations in memory and tipping the server into swap. On Cloud VPS M you can raise parallelism to 2 if your workflows fire often.<\/p>\n
Step 3: The Docker Compose File (n8n + Ollama + PostgreSQL + Caddy)<\/h2>\n
This Compose file runs four containers on a private network. Only Caddy publishes ports 80 and 443 to the internet. n8n and Ollama talk to each other by container name, and Ollama’s port 11434 is never exposed on the host, which is the single most important security decision in this whole setup.<\/p>\n
# docker-compose.yml\nservices:\n  postgres:\n    image: postgres:16-alpine\n    restart: unless-stopped\n    environment:\n      POSTGRES_USER: ${POSTGRES_USER}\n      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD}\n      POSTGRES_DB: ${POSTGRES_DB}\n    volumes:\n      - postgres_data:\/var\/lib\/postgresql\/data\n    healthcheck:\n      test: [\"CMD-SHELL\", \"pg_isready -U ${POSTGRES_USER} -d ${POSTGRES_DB}\"]\n      interval: 10s\n      timeout: 5s\n      retries: 5\n\n  ollama:\n    image: ollama\/ollama:latest\n    restart: unless-stopped\n    environment:\n      OLLAMA_HOST: 0.0.0.0\n      OLLAMA_KEEP_ALIVE: ${OLLAMA_KEEP_ALIVE}\n      OLLAMA_NUM_PARALLEL: ${OLLAMA_NUM_PARALLEL}\n      OLLAMA_MAX_LOADED_MODELS: ${OLLAMA_MAX_LOADED_MODELS}\n    volumes:\n      - ollama_models:\/root\/.ollama\n    # No \"ports:\" block on purpose. Ollama is reachable only from n8n.\n\n  n8n:\n    image: docker.n8n.io\/n8nio\/n8n:latest\n    restart: unless-stopped\n    depends_on:\n      postgres:\n        condition: service_healthy\n    environment:\n      N8N_HOST: ${DOMAIN}\n      N8N_PORT: 5678\n      N8N_PROTOCOL: https\n      N8N_EDITOR_BASE_URL: https:\/\/${DOMAIN}\/\n      WEBHOOK_URL: https:\/\/${DOMAIN}\/\n      N8N_ENCRYPTION_KEY: ${N8N_ENCRYPTION_KEY}\n      GENERIC_TIMEZONE: ${GENERIC_TIMEZONE}\n      TZ: ${TZ}\n      DB_TYPE: postgresdb\n      DB_POSTGRESDB_HOST: postgres\n      DB_POSTGRESDB_PORT: 5432\n      DB_POSTGRESDB_DATABASE: ${POSTGRES_DB}\n      DB_POSTGRESDB_USER: ${POSTGRES_USER}\n      DB_POSTGRESDB_PASSWORD: ${POSTGRES_PASSWORD}\n      N8N_RUNNERS_ENABLED: \"true\"\n      EXECUTIONS_DATA_PRUNE: \"true\"\n      EXECUTIONS_DATA_MAX_AGE: 168\n    volumes:\n      - n8n_data:\/home\/node\/.n8n\n\n  caddy:\n    image: caddy:2\n    restart: unless-stopped\n    ports:\n      - \"80:80\"\n      - \"443:443\"\n    environment:\n      DOMAIN: ${DOMAIN}\n      ACME_EMAIL: ${ACME_EMAIL}\n    volumes:\n      - .\/Caddyfile:\/etc\/caddy\/Caddyfile:ro\n      - caddy_data:\/data\n      - caddy_config:\/config\n\nvolumes:\n  postgres_data:\n  ollama_models:\n  n8n_data:\n  caddy_data:\n  caddy_config:\n<\/code><\/pre>\n
The Caddyfile is three lines. Caddy requests and renews the Let’s Encrypt certificate itself.<\/p>\n
# Caddyfile\n{\n  email {$ACME_EMAIL}\n}\n\n{$DOMAIN} {\n  reverse_proxy n8n:5678\n}\n<\/code><\/pre>\n
Bring the stack up and watch the logs until n8n reports it is listening and Caddy reports a certificate.<\/p>\n
docker compose up -d\ndocker compose logs -f caddy n8n\n<\/code><\/pre>\n
Open https:\/\/n8n.yourdomain.in<\/code>, create the owner account, and you are in. If you would rather use Traefik, the only change is replacing the Caddy service with a Traefik container that has a Let’s Encrypt resolver and adding router labels to the n8n service; the rest of the file is unchanged. We use Caddy here because there is less to get wrong. The n8n self-hosting guide<\/a> covers the n8n-only version of this stack if you want the simpler starting point.<\/p>\n
Step 4: Pull a Model into Ollama<\/h2>\n
Pull the model from inside the Ollama container; the weights land in the ollama_models<\/code> volume and survive restarts and updates. Match the model to your plan using the table above.<\/p>\n
# Cloud VPS S (8 GB): a 3B model is the safe default\ndocker compose exec ollama ollama pull llama3.2:3b\n\n# Cloud VPS M (16 GB): 7B\/8B at Q4 runs comfortably\ndocker compose exec ollama ollama pull llama3.1:8b\ndocker compose exec ollama ollama pull qwen2.5:7b\n\n# Check what is installed and test a prompt\ndocker compose exec ollama ollama list\ndocker compose exec ollama ollama run llama3.2:3b \"Reply with one word: ready\"\n<\/code><\/pre>\n
The first response after a pull or a restart takes longer because the model has to load into RAM; after that, OLLAMA_KEEP_ALIVE=30m<\/code> keeps it resident between workflow runs. Watch memory while the model answers: docker stats --no-stream<\/code> will show you whether you are near the plan’s limit. If you see swap use in free -m<\/code>, step down a model size or step up a plan. For a DeepSeek-family model instead, our DeepSeek on VPS guide<\/a> has the sizing notes.<\/p>\n
Step 5: Connect the n8n Ollama Node<\/h2>\n
n8n reaches Ollama over the Docker network at http:\/\/ollama:11434<\/code>, and that address goes into one credential that every Ollama node reuses. In the n8n editor: Credentials, Add credential, search for Ollama, set Base URL to http:\/\/ollama:11434<\/code>, save. The test button should return a green tick; if it fails, the two containers are not on the same Compose network or Ollama has not finished starting.<\/p>\n
There are two Ollama nodes in n8n and both are sub-nodes that attach to a parent:<\/p>\n