Setting Up Ollama with Microsoft Power BI

What is Ollama?

Why Use Power BI?

Advantages of Setting Up Ollama with Power BI

• Enhanced Data Control: Running Ollama locally means that your data remains within your organizational firewall, thus reducing the risk of data breaches often associated with cloud-based solutions.
• Improved Performance: Local model inference can cut down on latency and enhance response times compared to querying AI models hosted on remote servers.
• Cost Savings: Avoiding continuous subscription fees makes this combination very appealing, especially for startups and smaller organizations.
• Customization: You can easily tailor Ollama’s models to fit specific data needs and requirements, ensuring a more aligned use-case application.

Prerequisites

• A functional Microsoft Power BI account.
• A computer capable of running Ollama (preferably with good RAM & CPU).
• Docker installed on your system—the versatile container hosting platform.
• Basic understanding of RESTful APIs.

Step-by-Step Setup Guide

Step 1: Install Ollama

1. Go to the Ollama download page.
2. Choose the appropriate installation based on your operating system (Windows, macOS, or Linux).
3. Follow the installation instructions.

Step 2: Pull Ollama Docker Image

1. Open your terminal or command prompt.
2. Use the following command to pull the Ollama Docker image:
   1 2 bash docker pull ollama/ollama

Step 3: Run Ollama Using Docker

1. Start the Ollama container by executing:

   1 2 bash docker run -d --gpus=all -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama
   Note: The

   1 --gpus=all

   flag indicates you're utilizing GPUs for model inference, which is vital for speedy interaction.

Step 4: Load Your Models

1. Open a new terminal and execute:
   1 2 bash ollama run llama3
2. Wait for the download verification to confirm that Llama 3 is ready for use.

Step 5: Integrate with Power BI

1. Open Power BI Desktop.
2. Under Home, navigate to Get Data.
3. Choose Web to retrieve data through API requests.
4. Input the REST endpoint to access Ollama models. Here’s a sample URL to get you going:

   1 http://localhost:11434/api/chat
   Ensure your network allows the necessary connections to that endpoint.

Step 6: Make API Calls

1. Go to Transform Data in Power BI.
2. Under the Home tab, click on Advanced Editor.
3. Insert a query to use Ollama’s chat feature. Here's an example of what that might look like:
   1 2 3 4 5 text let Source = Json.Document(Web.Contents("http://localhost:11434/api/chat", [Headers = ["Content-Type"="application/json"], Content = Text.ToBinary("{ \"model\": \"llama3\", \"messages\": [{ \"role\": \"user\", \"content\": \"Hello! Tell me about the benefits of Power BI?\" }] }")])) in Source
   This sends a basic message to your Llama 3 model, and Power BI retrieves the generated response.

Step 7: Visualize the Data

1. Drag & drop visuals you want to deploy.
2. Customize the layout to fit your desired aesthetic.

Bonus: Using Arsturn for Enhanced Engagement

• No Coding Required: Perfect for everyone, even if you're a total newbie!
• Instant Responses: Ensure customers get the info they need, when they need it.
• Full Customization: You'll control the chatbot's personality, content, & appearance to fit your brand identity.

Conclusion