Hacker Timesnew | past | comments | ask | show | jobs | submitlogin

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



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

Search: