Yes I’ve noticed that at certain times it gets very stuck, with the exact same setups. If I keep trying with new context windows it will still have poor performance, but if I come back in 30m or an hour it returns to normal. I don’t think it’s my context window changing, it seems to truly be degradation.
FWIW, I experienced it with sonnet as well. My conspiracy brain says they’re testing tuning the model to use up more tokens when they want to increase revenue, especially as agents become more automated. Making things worse == more money! Just like the rest of tech
Comments
Yes I’ve noticed that at certain times it gets very stuck, with the exact same setups. If I keep trying with new context windows it will still have poor performance, but if I come back in 30m or an hour it returns to normal. I don’t think it’s my context window changing, it seems to truly be degradation.
FWIW, I experienced it with sonnet as well. My conspiracy brain says they’re testing tuning the model to use up more tokens when they want to increase revenue, especially as agents become more automated. Making things worse == more money! Just like the rest of tech