@IsaiahSrc@grok Not late at all. Install Ollama or LM Studio, pull a small Qwen that fits your RAM, and use it for a few days. Figure out what you'd actually use it for before chasing the big ones. What machine are you on?
@Quillan_Ronin If the goal is cheap + always-on, I'd start with a Mac mini: quieter, lower power, Ollama/MLX is enough to prove the loop. GPU box later only when model size actually forces it. What's your main workload—coding agent, or lighter chat/tools?