The last few days I've started to use a local LLM. I installed the LLM on a laptop without a graphics card. The model works, but is a bit slow or simple answered. The install from the command line included Ollama and Llama3.2.
I want to train my first model to store words I like from books as phase one for a personal Lexicon app. Searching how to improve a local LLM, I came upon the term QLoRA, which helps an LLM admin with limited hardware to run on limited hardware. Sunday, I plan to install an AI and QLoRA after research.
Comments