This happens when the model uses it’s full response budget during reasoning. Usually this is symptomatic of too broad of a prompt, or too high of a selected reasoning level. As GLM does not support variable reasoning levels, I suspect the former.
Can you share any more detail about what the prompt was that you most recently encountered this with?
Log in to comment and vote
Comments1
Mark Florkowski
This happens when the model uses it’s full response budget during reasoning. Usually this is symptomatic of too broad of a prompt, or too high of a selected reasoning level. As GLM does not support variable reasoning levels, I suspect the former.
Can you share any more detail about what the prompt was that you most recently encountered this with?