Towards AI Can Help your Team Adopt AI: Corporate Training, Consulting, and Talent Solutions.

Publication

How to build an Air-gapped LLM-based AI Chatbot in Containers Step-by-Step
Artificial Intelligence   Latest   Machine Learning

How to build an Air-gapped LLM-based AI Chatbot in Containers Step-by-Step

Last Updated on April 22, 2024 by Editorial Team

Author(s): Mélony Qin (aka cloudmelon)

Originally published on Towards AI.

As AI tools become increasingly popular, they play an important role in boosting our productivity in everyday tasks. However, bringing them into professional settings faces challenges because they need internet access, and some company policies simply do not allow them. The solution is not to break the company firewall but to build a real air-gapped AI and make it up and running locally or in containers.

So, in this blog post, I will share with you how to build an Air-gapped LLM-based AI Chatbot in Containers Step-by-Step by leveraging open-source technologies such as Ollama, Docker, and the Open WebUI.

Make AI learning simple and fun

Ollama is the simplest way I’ve found to set up and run large language models(LLMs) locally. With many years of experience working with containers and Kubernetes, I like to think we could apply similar philosophies to this technology. Let me show you why!

Similar to pulling container images from Docker Hub, with Ollama, it’s easy to pull AI models from Ollama’s model library to your local computer and run the Ollama command to interact with AI models directly. The overall comparison is like the hand-drawn diagram I’ve made:

How each layer works with ollama

Ollama isn’t the only option for this… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI

Feedback ↓