Look at a Small Language Models or Gemma4:e2b for generating comments on the server without paying an API for tokens. I explored running these locally for making a very unsophisticated chat bot and it was just slightly too slow for what I wanted, but you don't need instantaneous responses for generating comments on posts.
It's awesome! And it's always cool to me when I get a chance to look at a largerish CL codebase! Congratulations on all the hard work now mostly complete!
> It's an extremely efficient port of HN to Common Lisp. One of the project's goals was to be high performance.
FWIW your pages seem to load at least 2 times slower here. no idea if real HN has some sort of CDN cache though.
Look at a Small Language Models or Gemma4:e2b for generating comments on the server without paying an API for tokens. I explored running these locally for making a very unsophisticated chat bot and it was just slightly too slow for what I wanted, but you don't need instantaneous responses for generating comments on posts.
It's awesome! And it's always cool to me when I get a chance to look at a largerish CL codebase! Congratulations on all the hard work now mostly complete!
> If you're particularly offended that this is copying your comments, claim your account by
Does this comply with GDPR?
good job!