How to configure the Ollama serve port for network access

What this does

Rebinds the Ollama API server from localhost-only to a network-accessible address and port. After this guide remote clients will be able to reach the Ollama API over the network.

Steps

1. Set the OLLAMA_HOST environment variable and restart. Changes the bind address to accept external connections.

   export OLLAMA_HOST=0.0.0.0:11434
   ollama serve
   

   Expected output: Server starts, listening on all interfaces.

2. Configure the systemd service to persist the change. Adds the environment variable to the unit file.

   sudo mkdir -p /etc/systemd/system/ollama.service.d
   sudo tee /etc/systemd/system/ollama.service.d/override.conf << 'EOF'
   [Service]
   Environment=OLLAMA_HOST=0.0.0.0:11434
   EOF
   sudo systemctl daemon-reload && sudo systemctl restart ollama
   

3. Open the firewall port. Allows inbound TCP traffic on the Ollama port.

   sudo ufw allow 11434/tcp && sudo ufw reload
   

   Expected output: Rule added and Rules updated.

4. Test remote access. Sends an API request from another machine to confirm the service is reachable.

   curl http://<server-ip>:11434/api/tags
   

   Expected output: A JSON array of available models.

Verification

curl -s http://localhost:11434/api/tags | python3 -m json.tool
# Expected: valid JSON with non-empty "models" array

Common failures

• connection refused - Firewall blocking the port; verify with sudo ufw status or check cloud security groups.
• address already in use - Another process is on port 11434; use sudo lsof -i :11434 to identify it.
• OLLAMA_HOST not respected - The daemon may be started by systemd which ignores the shell export; always use the override.conf method for systemd setups.
• model unreachable after config - Client is using an IP that the server is not bound to; verify with ss -tlnp | grep ollama.

Operator checkpoint

Before treating this as solved, write down the local runtime, model or package version, hardware/backend if relevant, and the verification output. This keeps the guide useful as a Will-It-Run style decision instead of a one-off command transcript.

Related guides

• How to run a model in the background as a persistent service
• How to verify model integrity after downloading