@thdxr GLM-5.3-Flash is cheaper than DeePseek-V4-Flash, but it actually has fewer call counts in opencode compared to DeePseek-V4-Flash. This model is not large; why is that?
@opencode The number of calls dropped from 150,000 to 18,000. As far as I know, OpenCode has its own deployment, right? This change is too drastic and outrageous.