I was using GLM 5.1 version well, and when version 5.2 came out, I observed it for a few days and started using it from yesterday
For some reason, the usage is draining steadily, and from what I read somewhere, token usage increased by about 70% compared to 5.1
From my experience using it, it seems to be almost 3 times more????
Certainly, the context is larger than before, so it takes a bit longer to get polluted and it remembers things longer, but the problem is that there's so much token usage that I wonder if using this is even right
I think I need to go back to 5.1. For reducing context, I should resolve it by saving more frequently and starting a new session. I was using the previous version well anyway
I didn't change any settings, only changed the model, and this seems to have some kind of problem????
