1 Comment
User's avatar
Popular AI's avatar

Would you test GLM-5.2 through an API first, wait for better local AI quantizations, or try to run this open coding model yourself anyway?