I have a pc with 24GB 3090 card capable of running LLMs locally, but our electricity costs make API calls much more reasonable.
Even gaming -- streaming geforce now is cheaper than ammortizing cost of power + hardware over time
I have a pc with 24GB 3090 card capable of running LLMs locally, but our electricity costs make API calls much more reasonable.
Even gaming -- streaming geforce now is cheaper than ammortizing cost of power + hardware over time