I evaluated GLM 5.2 against the frontier on tasks from real repos(stet.sh)

2 pointsby bisonbearvor 21 Tagen2 comments

https://www.stet.sh/blog/glm-5-2-passes-tests-fails-review

2 comments

livalayvor 21 Tagen

Why did you use GLM 5.2 on medium when you used all the other models on high or xhigh?

re-thcvor 21 Tagen

What does that even map to? On the official GLM 5.2 blog it is none, high and max.