You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi!
I am new to the framework and I would like to understand how much control we have on the chat history.
All the APIs out there like Gemini, Claude, OpenAI etc, are based on a chat interface with a list of messages like Human, AI, Tool, etc.
I have a chat application where the user chats to the AI and at each turn send the list of messages to the LLM API which returns an output that is then appended to the history and the process continues like that. Essentially I am handling the chat history outside of the agent.
I know that CrewAI handle the memory internally or through databases for long term memory. Is there a way to pass the conversation memory too? So far I am thinking that in such scenario CrewAI agents should be used to solve a specific problem but I need to handle the conversation with the user in a different way where perhaps the crewAI agents are a tool of the outer agent. Are my assumption correct? Thanks!
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hi!
I am new to the framework and I would like to understand how much control we have on the chat history.
All the APIs out there like Gemini, Claude, OpenAI etc, are based on a chat interface with a list of messages like Human, AI, Tool, etc.
I have a chat application where the user chats to the AI and at each turn send the list of messages to the LLM API which returns an output that is then appended to the history and the process continues like that. Essentially I am handling the chat history outside of the agent.
I know that CrewAI handle the memory internally or through databases for long term memory. Is there a way to pass the conversation memory too? So far I am thinking that in such scenario CrewAI agents should be used to solve a specific problem but I need to handle the conversation with the user in a different way where perhaps the crewAI agents are a tool of the outer agent. Are my assumption correct? Thanks!
Beta Was this translation helpful? Give feedback.
All reactions