4 ms·

What about Kimi and GLM?

• zozbot234 8mo ago
  These are well behind the general state of the art (1yr or so), though they're arguably the best openly-available models.
  • tgrowazay 8mo ago
    According to artificial analysis ranking, GLM-5 is at #4 after Claude Opus 4.5, GPT-5.2-xhigh and Claude Opus 4.6 .
  • epolanski 8mo ago
    Idk man, GLM 5 in my tests matches opus 4.5 which is what, two months old?
    • wahnfrieden 8mo ago
      4.5 was never sota