Post #2721513

@gruber From what you said about how much and how you use chatGPT you'd probably be equally or better served running a local model. Amusingly, chatGPT (or Gemini, what I used) would make it really easy for you to set up lmstudio and just never pay for a subscription ever again. (I'm running smaller models on my M1 Mac Mini, and it's not blazingly fast, but it does proofreading and small-scale code generation really well.)