Hacker News new | ask | show | jobs
by slagfart 44 days ago
GLM5.2 isn't even on this benchmark
1 comments
ttul 43 days ago
True. Z.AI ran that bench themselves and report 46.2, which is lower than GPT-5.5 and Opus 4.8, but crushing the other open weights models.

https://z.ai/blog/glm-5.2

link