3 Comments

  1. Like as im a cse student i did try to run a model , but due to the my low end specs the response was slow.
    But if i build a capable pc to run the ai model with higher info and then integrating that to a self made app for taking and sending messages , maybe that could help me in saving the pro model costs

  2. Spara-Extreme on

    AI is a feature, not a full blown product. Local models will absolutely deliver most of what ChatGPT and Gemini do relatively soon but that’s not going to net in any type of cost savings. The costs will shift to something else- whether that’s specialized hardware or increases in price for other “AI powered” services.

    It’s better to think of the benefits of local models being that you won’t lose as much data to privacy violations as you would with cloud offerings.

  3. Cheapskate-DM on

    The biggest advantage of local models is the lack of drift from the Big Model poisoning itself with AI-generated web content and corporate twiddling. Fixed control over the inputs should have always been the standard.