What are Large Language Models?
To understand large language models(LLMs) you have to understand that fundamentally LLMs take in an input, or combination of inputs, and produce an output. They create this output from entirely within their own self contained structure. A LLM on its own is kind of like a brain, generating text, audio, images, etc… in its mind and feeding it to a user through a visual display.
There are 3 main distinct types of Large Language Models for handling different types of generation: Image, Text and Audio
There are also types of LLMs that handle generating multiple types of outputs, these are called MultiModal Models as they can handle multiple modalities of input
For example, GPT4 can receive and generate text and images making it a multimodal model
Where is all of the application's data stored?
All of your data( data you create within the application, or any downloaded large language models and any data created by them) is processed and stored on your local computer. The only internet connection necessary is the authentication process
Does any of my data leave my computer?
No
Why use Large Language Models?
Large language models are useful because they make getting information and completing tasks easier and faster. They can help you write emails, come up with ideas, or answer questions you're curious about. They're like having a super knowledgeable assistant who's always available. You can use them for learning, solving problems, or just for fun conversations. They're especially handy because they can understand and communicate in a way that feels natural and human-like.