Is your feature request related to a problem? Please describe.
In a story writing context, it is important to maintain the context window so would not trigger max token limit.
Describe the solution you'd like
- In a long text generation scene the autogen can automatically handle the context window. Auto summarize it when it reaches the max token limitation.
- Provide functions to control which message should be kept in memories. Can defined the message priorities.
Additional context
No response
Is your feature request related to a problem? Please describe.
In a story writing context, it is important to maintain the context window so would not trigger max token limit.
Describe the solution you'd like
Additional context
No response