reminix.adapters.llamaindex.from_chat_engine¶

reminix.adapters.llamaindex.from_chat_engine(chat_engine, *, name, metadata=None)[source]¶

Create a Reminix Agent from a LlamaIndex ChatEngine.

ChatEngine maintains conversation history and is ideal for conversational RAG applications.

Parameters:

chat_engine (LlamaIndexChatEngineProtocol) – A LlamaIndex ChatEngine instance.
name (str) – Name for the Reminix agent.
metadata (dict[str, Any] | None) – Optional metadata for the agent.

Return type:

Agent

Returns:

A Reminix Agent that wraps the ChatEngine.

Example:

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
from reminix.adapters.llamaindex import from_chat_engine
from reminix.runtime import serve

# Load and index documents
documents = SimpleDirectoryReader("data").load_data()
index = VectorStoreIndex.from_documents(documents)

# Create ChatEngine and wrap it
chat_engine = index.as_chat_engine(chat_mode="condense_question")
agent = from_chat_engine(chat_engine, name="docs-chat")

serve(agent)