Speakers
Description
Artificial Intelligence and large language models are most frequently associated with Open AI, Google Bard, or Microsoft CoPilot (Bing). Most of these AIs are paid services; however, building and running your own AI is surprisingly simple. Moreover, there is no need for expensive server hardware; it is possible to run your own private AI within your laptop or even on a Raspberry Pi. Running your own private AI comes with many advantages. For example, you can choose exactly which model(s) to make available to users and how much computing power to allocate to it; being self-hosted and owned, you retain control of your data and are not giving it away to corporations; you retain complete freedom to use the software and your hardware in any way you choose. This poster will present Ollama, a simple, free, open-source platform for launching and running AI models. Installing on your devices is as simple as running a single command or launching a docker container. There are several open-source models available that have been trained for a variety of different purposes. Adding a web user interface can also be done, connecting to the Ollama API for your own private "ChatGPT" without needing so much as an external internet connection. Participants will also be able to try out the AI as we will bring one along with us to demonstrate.
| Keywords | AI, open-source, large language model |
|---|
