Recently, I found myself looking into options for running local AI models which leveraged my home lab. The adoption of Ollama is heavily prevalent in many opensource options as an integration point. Jan continues to be developed and tweaked to support model parameter tuning with an incredibly straight forward GUI.