Half the “problems” have been solved by retro hardware enthousiasts. Acting like it’s a big deal you have to connect a mouse and keyboard to the ps/2 ports. Even back in the early 2000s this was solved with USB to ps2 dongles.

The 1B parameter version of Llama 3.2 showed even slower results at 0.0093 tokens per second, based on the partial model run with data stored on disk.

I mean, cool? They got a C interface library to compile using an older C standard, and the 1B model predictably runs like trash. It will take hours to do anything meaningful at that rate.