GLM 5.2: how to set up local AI (with Cursor/Codex) 1. How to actually set it up in Cursor and Codex (most people are overcomplicating this). 2. Why local models suddenly matter way more than they did a month ago. 3. The cost difference between running this and running Opus. It's not close. 4. The model chaining trick that gives you the best output for a fraction of the price. 5.Does it make sense to buy hardware? The honest answer. Full ep on today's ep of @startupideaspod.…