Reimage Your Server for Ollama Installation

Ollama is a tool that lets you run large language models (LLMs) locally on your own computer, instead of relying on cloud-based AI services. Ollama makes it easy to download, run, and interact with AI models, such as chatbots, entirely offline.

How Ollama works

• Runs AI models locally (Mac, Linux, Windows support)
• No internet required after model download
• Simple command-line interface
• Privacy-friendly (your data stays on your machine)
• Optimized for consumer hardware

What models can Ollama run?

Ollama supports many popular open-source LLMs, for example:

• Llama 2 / Llama 3
• Mistral
• Gemma
• Phi
• Code-focused models (for programming help)

System Requirements

Before deploying, ensure your server meets the requirements below. For a stable production deployment, we recommend using higher specifications than the Approximate Size requirements.

Ollama AI Model Storage Requirements

Ensure your server has sufficient disk space available to download and store the specific AI models you plan to run locally in Ollama:

How to Reimage Your Server for Ollama Installation

1. Log in to your Bluehost Portal.
2. In the left-hand menu, click Hosting, then select My Plans.

   
    

3. Click the Manage button on the hosting package you want to reimage.

   
    

4. Navigate to the Server Image section, then click the Reimage button.

   
    

5. Select the Applications tab.

   
    

6. From the list of available applications, find Ollama and click Select.

   
    

7. On the pop-up message, type reimage in the field, then click Proceed.

   
    

8. Wait a few minutes while the installation completes.

   
    

9. Once completed, you will see Ollama listed in the Server Image section.

   
    

How to Use Ollama

After the installation completes, follow the steps below to set up your application.

1. Log in to your Bluehost Portal.
2. In the left-hand menu, click Hosting, then select My Plans.

   
    

3. Locate the server you want to manage, then click View Details in the bottom-left corner to expand the section and access additional server information.

   
    

4. Click the Reset Password.

   
    

5. In the pop-up window, type your new root password, then click Reset Password again to confirm and save the changes.

   
    

6. Connect to your VPS via SSH.
   Open your computer's terminal (or an SSH client like PuTTY) and run the following command to log into your server as the administrator, using your server's public IP address:
   Replace your_server_ip with your actual server IP address.

   ssh root@your_server_ip

   Follow the prompts and enter the new root password you created in the step above.

   Example Output:

   
   Welcome to Ubuntu 24.04.4 LTS (GNU/Linux 6.8.0-111-generic x86_64)
   ********************************************************************************
   Ollama is running on this server.
   
   The UFW firewall is enabled.
   All ports are BLOCKED except 22 (SSH) and 11434 (Ollama API).
   
   API endpoint:
     http://localhost:11434
   
   Common commands:
     ollama list                  # list downloaded models
     ollama pull llama3.2         # download a model (~2GB)
     ollama pull tinyllama        # download a small starter model (~637MB)
     ollama run tinyllama         # start an interactive chat session
   
   REST API examples:
     curl http://localhost:11434/api/version
     curl http://localhost:11434/api/tags
     curl -X POST http://localhost:11434/api/generate \
          -d '{"model":"tinyllama","prompt":"Hello","stream":false}'
   
   On the server:
     Models are stored in /var/lib/ollama/.ollama/models
     Service config: /etc/systemd/system/ollama.service.d/override.conf
   
   For more details please see /root/README.md
   
   To delete this message of the day: rm -rf /etc/update-motd.d/99-ollama
       

7. Check installed and available models via the CLI.
   Copy

   ollama list
   Example Output:

   
   NAME        	ID          	SIZE    MODIFIED
   tinyllama   	sha256:e5562a3b…	637 MB  3 minutes ago
       

8. Pull a model (tinyllama is preinstalled on first boot):
   Copy

   ollama pull llama3.2
   Example Output:

   
   pulling manifest
   pulling 2.0 GB model
   ...
   success
       

9. Start an interactive session with a model:
   Copy

   ollama run tinyllama
   Example Output:

   >>> Hello, how can I help you today?
   (type /exit to quit)
   

10. Remove a model to free up disk space:
    Copy

    ollama rm tinyllama
    Example Output:

    
    Deleted model 'tinyllama'
        

11. Interact with the Ollama REST API
    • Check Ollama version:
      Copy

      curl http://localhost:11434/api/version
      Example Output:

      
      {"version": "0.1.25"}
              

    • List available models (tags):
      Copy

      curl http://localhost:11434/api/tags
      Example Output:

      
      {"models":[{"name":"tinyllama","modified_at":"2024-06-18T10:17:22.093Z"}]}
              

    • Generate a non-streaming completion:
      Copy

      curl -X POST http://localhost:11434/api/generate \ -H 'Content-Type: application/json' \ -d '{"model":"tinyllama","prompt":"Why is the sky blue?","stream":false}'
      Example Output:

      
      {"response":"The sky looks blue because of the way sunlight interacts with Earth's atmosphere...","model":"tinyllama",...}
              

    • Start a chat session via API:
      Copy

      curl -X POST http://localhost:11434/api/chat \ -H 'Content-Type: application/json' \ -d '{"model":"tinyllama","messages":[{"role":"user","content":"Hello"}]}'
      Example Output:

      
      {"message":{"role":"assistant","content":"Hello! How can I help you?"}, ...}
              

12. Model storage location
    Models are stored in
    Copy

    /var/lib/ollama/.ollama/models

    Before pulling large models, make sure you have sufficient disk space (7B models may need 4-8 GB free space).

13. Service config and logs
    • Service configuration is at
      Copy

      /etc/systemd/system/ollama.service.d/override.conf
    • For logs or troubleshooting, see:
      Copy

      cat /root/README.md

Links

• Ollama Documentation

Additional Information and Example Documentation Output

You can run cat /root/README.md to see more information and useful commands:

Example Output:

root@12-1-1-123:~# cat /root/README.md
# Ollama

## Description

Ollama is a platform that lets you run and manage AI models locally on
your server with full control and privacy. It makes it easy to
download, deploy, and interact with large language models, enabling
you to build AI-powered applications, automate tasks, and process data
without relying on external services.


## Services

Service    | Ports   | Purpose
-----------|---------|---------
ollama API | 11434   |
OpenSSH    | 22      | SSH

Ports are protected using ufw. Port 11434 is open to allow external API access.


## How-to-use

On first boot the image will automatically pull **tinyllama** (~637 MB) as a starter
model so the API is ready to use without any manual steps.

### Interacting via CLI

SSH into the server and use the `ollama` command:

```bash
ollama list                    # list downloaded models
ollama pull llama3.2           # pull a model (~2 GB)
ollama pull tinyllama          # pull a small model (~637 MB)
ollama run tinyllama           # start an interactive session
ollama rm tinyllama            # remove a model
```

### Interacting via REST API

The API is accessible from outside the server on port 11434:

```bash
# Check version
curl http://localhost:11434/api/version

# List available models
curl http://localhost:11434/api/tags

# Generate a completion (non-streaming)
curl -X POST http://localhost:11434/api/generate \
     -H 'Content-Type: application/json' \
     -d '{"model":"tinyllama","prompt":"Why is the sky blue?","stream":false}'

# Chat endpoint
curl -X POST http://localhost:11434/api/chat \
     -H 'Content-Type: application/json' \
     -d '{"model":"tinyllama","messages":[{"role":"user","content":"Hello"}]}'
```

### Model storage

Models are stored in `/var/lib/ollama/.ollama/models`. Ensure your volume has
sufficient space before pulling large models (7B parameter models are typically 4-8 GB).


## Links

Documentation: [https://docs.ollama.com/](https://docs.ollama.com/)

Summary

Ollama becomes even more powerful when paired with a Bluehost Self-Managed VPS or Virtual Dedicated Server (VDS). Running models on a VPS or VDS gives you dedicated resources, full control over configuration, and the flexibility to scale beyond a local machine—while still keeping your AI environment private. With Ollama deployed on a Self‑Managed VPS or VDS, teams can experiment, prototype, or support internal tools with consistent performance and greater reliability. Together, Ollama and a Bluehost VPS or VDS create a practical path to owning your AI stack—secure, customizable, and built on infrastructure you control.