Hacker News new | ask | show | jobs
by nkozyra 606 days ago
Probably faster to use off the shelf model with llama.cpp than to mock it