[HN] Ask HN: Who is using small OS LLMs in production?

1 crosspost

TechTakes @awful.systems my chatbot is so efficient it only needs one $2000 GPU per user