Local models are a huge feature. Using LMX models in LMStudio is faster than ollama in macos sillicon, so please enable server connection with LM Studio