Stateful Agents: Memory for Language Models
from The Gist Talk Hosted by kw
The YouTube workshop, led by Charles Packer, explores the concept of stateful agents, emphasizing the importance of memory in moving beyond current stateless LLM-driven agents. Packer introduces MemGPT as a memory management system for AI, contrasting it with simpler methods of appending to context windows. The workshop's practical component involves using Docker and a Jupyter notebook to interact with a server demonstrating stateful agents. Furthermore, the discussion covers the Leta framework, a server-client architecture designed for building persistent, stateful agents with features like core and archival memory, tool usage, and multi-agent communication, as showcased through a UI and a collaborative example
Embed this episode
No transcript yet.