How to Run Llama 3 8b with Ollama and Build AI Apps with Anakin AI

Llama 3 is the latest large language model (LLM) developed by Meta, available in 8 billion and 70 billion parameter sizes. It represents a significant advancement in natural language processing capabilities, outperforming many open-source chat models on common benchmarks. In this comprehensive guide, we'll explore how to run Llama 3 8b locally using Ollama and build AI apps with Anakin AI, a no-code platform for creating customized AI applications.

Running Llama 3 8b with Ollama

Ollama is a free and open-source application that allows you to run various large language models, including Llama 3, on your local machine, even with limited resources. It takes advantage of the performance gains of llama.cpp, an open-source library designed to enable running LLMs locally with relatively low hardware requirements.

Step 1: Install Ollama

The first step is to install Ollama on your system. It supports all major operating systems, including Windows (in preview), macOS, and Linux. Visit the Ollama website (https://ollama.com/) and download the appropriate version for your platform.

Step 2: Run Llama 3 8b

Once Ollama is installed, open your terminal or command prompt and run the following command to start Llama 3 8b:

ollama run llama3:8b

This command will download and load the 8 billion parameter version of Llama 3. Wait a few minutes while the model is downloaded and loaded, and then you'll be presented with a chat prompt similar to this:

ollama run llama3:8b

>>> Who was the second president of the United States?

The second President of the United States was John Adams. He served from 1797 to 1801, succeeding George Washington and being succeeded by Thomas Jefferson.

>>> Who was the 30th?

The 30th President of the United States was Calvin Coolidge! He served from August 2, 1923, to March 4, 1929.

>>> /bye

You can now interact with Llama 3 8b by typing your queries or prompts, and the model will generate responses based on its training.

Step 3: Explore Ollama's Features (Optional)

Ollama offers several additional features that you can explore:

Open WebUI: Open WebUI is an extensible, self-hosted UI that runs entirely inside Docker. It provides a user-friendly interface for interacting with Ollama and other OpenAI-compatible LLMs. You can install and run Open WebUI alongside Ollama for a more intuitive experience.

Ollama API: Ollama provides an API that allows you to integrate LLMs into your own projects. You can use the Ollama API or the OpenAI-compatible API to make requests and receive responses from the loaded LLM.

Custom Model Files: Ollama allows you to create custom model files with specific prompts, parameters, and behaviors. This feature enables you to fine-tune the LLM's behavior to suit your specific needs.

Model Management: Ollama provides commands for pulling, pushing, and managing different LLM models on your local system, giving you flexibility in choosing the models you want to work with.

Building AI Apps with Anakin AI

Anakin AI is a no-code platform that allows you to create customized AI apps without any coding knowledge. It offers a range of pre-built AI applications that you can customize according to your needs, as well as the ability to create new apps from scratch.

Step 1: Sign Up for Anakin AI

Visit the Anakin AI website (https://www.anakin.ai/) and sign up for a free account. Once logged in, you'll have access to your personal workspace.

Step 2: Choose a Pre-built App or Create a New One

Anakin AI offers a library of pre-built AI apps that cover various industries and use cases. For example, if you're looking to generate YouTube short scripts, you can select the pre-made app for that purpose.

Alternatively, you can create a new app by clicking on the "Add or Create an App" button in your workspace. Select the "Quick App" option and name your app.

Step 3: Customize the App

Once you've chosen or created an app, you can customize it based on your specific needs. Define the input fields, such as the topic for the YouTube short script or the desired output format. You can also choose the AI model you want to use and adjust other parameters.

Step 4: Generate Output

After customizing the app, click the "Generate" button to generate the desired output. Anakin AI will use the selected AI model and your input parameters to generate the content or perform the specified task.

Step 5: Explore Advanced Features (Optional)

Anakin AI offers several advanced features that you can explore:

API Integration: Anakin AI provides comprehensive API services for all applications, empowering developers and organizations to seamlessly integrate and enhance their projects using Anakin AI's APIs. By leveraging these APIs, users gain the flexibility to easily access Anakin AI's robust product features within their own applications.

To use the API, you'll need to upgrade your plan, generate an API access token, and follow the provided documentation for making API requests. Anakin AI's API integration allows you to rapidly develop AI applications tailored to your business needs, support multiple AI model providers, and stay ahead of the curve with upcoming advanced features.

Here's an example of how to use the API to create text completion:

curl --location --request POST 'https://api.anakin.ai/v1/quickapps/{{appId}}/runs' \
--header 'Authorization: Bearer ANAKINAI_API_ACCESS_TOKEN' \
--header 'X-Anakin-Api-Version: 2024-05-06' \
--header 'Content-Type: application/json' \
--data-raw '{
    "inputs": {
        "Product/Service": "Cloud Service",
        "Features": "Reliability and performance.",
        "Advantages": "Efficiency",
        "Framework": "Attention-Interest-Desire-Action"
    },
    "stream": true
}'

Replace {{appId}} with your app ID and ANAKINAI_API_ACCESS_TOKEN with your API access token.

Sharing Apps: If you find your customized AI apps useful, you can publish them on Anakin AI's App Store and share them with others using the provided link.

Pre-made Apps: Anakin AI offers a library of pre-made AI apps that cover various industries and use cases. These apps can be customized according to your specific needs, providing more refined results compared to building from scratch.

By combining the power of Llama 3 8b running locally with Ollama and the no-code capabilities of Anakin AI, you can unlock a world of possibilities for leveraging advanced natural language processing in your projects, without the need for extensive coding or technical expertise.

Building Workflow-Type AI Apps with Anakin AI

Anakin AI's Workflow feature allows you to create powerful, multi-step AI apps without any coding knowledge. Here are three examples of workflow-type AI apps you can build:

1. Internet-Connected ChatGPT

This app combines the power of ChatGPT with live internet search results, providing you with up-to-date and accurate information.

1. Define two variables: "Google Search" (single-line text) and "GPT Prompt" (long text).
2. Add the "Google Search" utility step and specify the "Google Search" variable as the search query.
3. Add an "AI Text Generator" step and craft a prompt instructing the AI to process the search results and present them in a structured format.
4. Set the output to display the AI's response.

When you test the app, it will perform a Google search based on your input, process the results using the AI model, and provide a detailed, well-formatted response.

2. Real-Time Currency Checker (API)

This app allows you to check real-time exchange rates by integrating with a currency API.

1. Define three variables: "Source Currency," "Target Currency," and "What are your needs?" (all single-line text).
2. Add the "API Request" utility step, set the method to POST, and enter the API URL and access key.
3. Add an "AI Text Generator" step and create a prompt introducing the AI as an exchange rate expert.
4. Set the output to display the AI's response.

When you test the app, it will fetch the current exchange rate from the API and provide a tailored response based on your input and the AI model's instructions.

3. Smart Art Generator

This app generates stunning AI-generated images from simple text descriptions by handling the complex prompt engineering behind the scenes.

1. Define a variable called "About the Art" (single-line text) for the user's text description.
2. Add an "AI Text Generator" step and craft a prompt instructing the AI to act as a proficient prompt crafter for visual art.
3. Add an "AI Image Generator" step and specify the output from the previous step as the prompt.
4. Set the output to display the generated image in Markdown format.

When you test the app, it will generate a high-quality image based on your text description, thanks to the AI's prompt engineering capabilities.

These examples demonstrate the versatility of Anakin AI's Workflow feature, allowing you to create powerful, multi-step AI apps tailored to your specific needs, without any coding knowledge.

By combining Ollama's local LLM capabilities with Anakin AI's no-code app-building platform and API integration, you can unlock a world of possibilities for leveraging advanced AI technologies in your projects, streamlining workflows, and enhancing productivity.