How to Easily Share Ollama API and Open WebUI Online

In today’s AI-driven world, deploying large language models (LLMs) like Meta’s Llama 3, Google’s Gemma, or Mistral locally offers unparalleled control over data privacy and customization. However, sharing these tools securely over the internet unlocks collaborative potential—whether you’re a developer showcasing a prototype, a researcher collaborating with peers, or a business integrating AI into customer-facing apps.

This comprehensive guide will walk you through exposing Ollama’s API and Open WebUI online using Pinggy, a powerful tunneling service. You’ll learn to turn your local AI setup into a globally accessible resource—no cloud servers or complex configurations required.

Why Share Ollama and Open WebUI Online?

The Rise of Local AI Deployment

With growing concerns about data privacy and API costs, tools like Ollama and Open WebUI have become essential for running LLMs locally. However, limiting access to your local network restricts their utility. By sharing them online, you can:

• Collaborate remotely with team members or clients.
• Integrate AI into web/mobile apps via Ollama’s API.
• Demo projects without deploying to the cloud.
• Reduce latency by keeping inference local while enabling remote access.

Why Choose Pinggy for Tunneling?

Pinggy simplifies port forwarding by creating secure tunnels. Unlike alternatives like ngrok, it offers:

• Free HTTPS URLs with no signup required.
• No rate limits on the free tier.
• SSH-based security for encrypted connections.

Prerequisites for Sharing Ollama and Open WebUI

A. Install Ollama

1. Download Ollama: Visit ollama.com and choose your OS:
   • Windows: Double-click the .exe installer.
   • macOS/Linux: Run

   curl -fsSL https://ollama.com/install.sh | sh 
   

2. Verify Installation:

   ollama --version  # Should return "ollama version 0.1.30" or higher
   

   

B. Download a Model

Ollama supports 100+ models. Start with a lightweight option:

ollama run qwen:0.5b  

For multimodal capabilities (text + images), try llava or bakllava:

ollama run llava:13b

C. Install Open WebUI

Open WebUI provides a ChatGPT-like interface for Ollama. Install via Docker:

docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main

Access the UI at http://localhost:3000 and create an admin account.

Step-by-Step Guide to Sharing Ollama API

Start Ollama Locally

By default, Ollama runs on port 11434. Start the server:

ollama serve  # Keep this terminal open

Create a Public URL with Pinggy

Use this SSH command to tunnel Ollama’s API:

ssh -p 443 -R0:localhost:11434 -t qr@a.pinggy.io "u:Host:localhost:11434"

Command Breakdown:

• -p 443: Connects via HTTPS for firewall compatibility.
• -R0:localhost:11434: Forwards Ollama’s port to Pinggy.
• qr@a.pinggy.io: Pinggy’s tunneling endpoint.
• u:Host:localhost:11434: Maps the tunnel to your local port.

After running, you’ll see a public URL like https://abc123.pinggy.link.

Test and Integrate the Shared API

Verify access using curl or Verify using web browser:

curl https://abc123.pinggy.link/api/tags

To test the Ollama API using JavaScript, follow these simple steps:

1. Clone the repository from GitHub: RunOllamaApi.
2. Open the project directory in your terminal.
3. Install the required dependencies by running npm install.
4. Execute the script with node main.js to test the API.

Step-by-Step Guide to Sharing Open WebUI

Expose Open WebUI via Pinggy

Run this command to share port 3000:

ssh -p 443 -R0:localhost:3000 a.pinggy.io

ssh -p 443 -R0:localhost:3000 a.pinggy.io

You’ll receive a URL like https://xyz456.pinggy.link.

Access the WebUI Remotely

1. Open https://xyz456.pinggy.link in a browser.
2. Log in with your Open WebUI credentials.
3. Leverage features like:
   • Chat with models (Llama 3, Mistral, etc.).
   • Upload documents for RAG (Retrieval-Augmented Generation).
   • Switch models via the top-right dropdown.

Advanced Configuration and Security Best Practices

Secure Your Deployment

• Add Basic Authentication to Pinggy:
  Append a username/password to your SSH command:

  ssh -p 443 -R0:localhost:3000 user:pass@a.pinggy.io
  

Custom Domains and Performance Optimization

Upgrade to Pinggy Pro (INR 204.89/month) for custom domains:

ssh -p 443 -R0:localhost:3000 -T yourdomain.com@a.pinggy.io

Real-World Use Cases for Remote AI Access

Collaborative Development

Distributed teams can:

• Share a single Ollama instance for code review and documentation generation.
• Train custom models collaboratively using Open WebUI’s RAG features.

Customer-Facing Applications

Expose Ollama’s API to power:

• AI chatbots for 24/7 customer support.
• Content generation tools for blogs, social media, or product descriptions.

Academic and Research Workflows

Researchers can securely share access to proprietary models with peers without exposing internal infrastructure.

Troubleshooting Common Issues

Connection Errors

1. “Connection Refused”:
   • Ensure Ollama is running: ollama serve.
   • Check firewall settings for ports 11434 and 3000.

Model Loading Failures

• Verify model compatibility with your Ollama version.
• Free up RAM for larger models like llama3:70b (requires 40+ GB).

Conclusion and Next Steps

By combining Ollama, Open WebUI, and Pinggy, you’ve created a secure, shareable AI platform without relying on cloud services. This setup is ideal for startups, researchers, or anyone prioritizing data control.