Comment by Computer0

4 hours ago

My usage of GLM 5.2 has been defined by slow throughput and flaky providers, in many (definitely not all!) applications a dumber, faster, and more consistent model makes more sense to me.

0 comments