LLM Agents

An agent is used to extend the functionality of a Large Language Model or Chat Model with the use of additional nodes such as triggers, retrievers, loaders and output parsers. Agent nodes are distinct from Chain nodes in that they automatically determine the correct attached node to use for a given user input.

Before you begin

• Learn about pathways
• Learn about pathway nodes

What agents are available for my pathway?

Ensemble supplies five types of Agent:

Agent	Description	Additional information
Conversation (default)	The Conversation agent is optimized for end-user conversation, and takes inputs from an attached chat node and chooses the appropriate model, memory or tool node to form its response.	Learn about the conversation agent
OpenAi Functions	Use functions available in your OpenAi account	Learn about the OpenAi functions agent
Plan and execute	Plan what to do based on inputs from compatible nodes, then execute tasks	Learn about the plan and execute agent
ReAct	A logical reasoning and action processing agent	Learn about the ReAct agent
SQL	Recive SQL inputs from compatible nodes to query a SQL data source	Learn about the SQL agent
Tools	Unified tool selection agent	Learn about the Tools agent