Typically you need about 1GB graphics RAM for each billion parameters (i.e. one byte per parameter). This is a 405B parameter model. Ouch.

Edit: you can try quantizing it. This reduces the amount of memory required per parameter to 4 bits, 2 bits or even 1 bit. As you reduce the size, the performance of the model can suffer. So in the extreme case you might be able to run this in under 64GB of graphics RAM.

Which, will also hurt the call center manager, until they hire enough staff to answer the fucking phones in less than 30 minutes.

It’s all software, even the stuff on the graphics cards. Those are the rasterisers, shaders and so on. In fact the graphics cards are extremely good at running these simple (relatively) programs in an absolutely staggering number of threads at the same time, and this has been taken advantage of by both bitcoin mining and also neural net algorithms like GPT and Llama.