Comment by Abishek_Muthian

I'm more interested in running distributed inference for purpose built small language models than these coding LLMs.

Say a distributed inference for image processing, SDR, local weather monitoring etc. These will run on mediocre specs and produce dependable output.

Nicely done OP.