- This memory allows for storing of messages and then extracts the messages in a variable.
- This memory keeps a list of the interactions of the conversation over time. It only uses the last K interactions
- This memory keeps a buffer of recent interactions in memory, and uses token length rather than number of interactions to determine when to flush interactions.
- This memory creates a summary of the coversation over time.
- Vector data memory
- Stores text (from conversation or elsewhere) in a vector database and retrieves the most relevent blocks of text.
- Entity memories
- Using an LLM, it remembers details about specific entities
- You can also use multiple memories at one time. Eg. conversation memory + entity memory to recall individuals
- You can also store the conversation in a conventional database (such as key-value store or SQL)