Show HN: I reduced LLM inference GPU calls by 94% using semantic routing(icomnewtechnologies.com)
2 pointsby kanacki8 hours ago1 comment
  • slach4 hours ago
    better publish it on github