▲ | jononor 5 days ago | |
ChatGPT and Claude do the same. And I have noticed that model performance can often degrade a lot before such a hard limit. So even when not hitting the hard limit, splitting out to a new session can be useful. Context management is the new prompt engineering... |