Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Are you using a smaller LLM for anything?
9 points by emschwartz 12 days ago | hide | past | favorite | 1 comment
Have you had any luck using a smaller model (<= 3B parameters) for anything? Every time I've poked around with them, they seem to stupid to follow the instructions I try to provide.

Curious if others have had any more luck and, if so, which model and for what use case.






I'm currently using a 32B Qwen model for summarization and it's pretty good.



Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: