Implementation:Elevenlabs Elevenlabs python RagConfig

Field	Value
source	Elevenlabs_Elevenlabs_python
domains	Conversational AI, RAG, Knowledge Retrieval, Embeddings
last_updated	2026-02-15

Overview

Description

RagConfig is a Pydantic model that defines the Retrieval-Augmented Generation (RAG) configuration for a conversational AI agent in the ElevenLabs platform. It controls how knowledge base documents are retrieved and used during conversations, including the embedding model, vector distance thresholds, document chunk limits, and custom query rewriting prompts.

This model is auto-generated by Fern from the ElevenLabs API definition and inherits from UncheckedBaseModel. It enables fine-grained control over the RAG pipeline to optimize relevance and performance of knowledge-grounded responses.

Usage

RagConfig is used within PromptAgentApiModelInput to configure how an agent retrieves and utilizes knowledge base content during conversations. It is essential for agents that need to answer questions based on uploaded documents or knowledge bases.

Code Reference

Source Location

src/elevenlabs/types/rag_config.py

Class Signature

class RagConfig(UncheckedBaseModel):
    ...

Import Statement

from elevenlabs.types import RagConfig

I/O Contract

Field	Type	Required	Description
`enabled`	`Optional[bool]`	No	Whether RAG is enabled for the agent.
`embedding_model`	`Optional[EmbeddingModelEnum]`	No	The embedding model to use for vector search.
`max_vector_distance`	`Optional[float]`	No	Maximum vector distance of retrieved chunks.
`max_documents_length`	`Optional[int]`	No	Maximum total length of document chunks retrieved from RAG.
`max_retrieved_rag_chunks_count`	`Optional[int]`	No	Maximum number of RAG document chunks to initially retrieve from the vector store. These are then further filtered by vector distance and total length.
`query_rewrite_prompt_override`	`Optional[str]`	No	Custom prompt for rewriting user queries before RAG retrieval. The conversation history will be automatically appended at the end. If not set, the default prompt will be used.

Usage Examples

Basic RAG Configuration

from elevenlabs.types import RagConfig

rag = RagConfig(
    enabled=True,
    max_documents_length=5000,
    max_vector_distance=0.8,
)

Advanced RAG with Custom Query Rewriting

from elevenlabs.types import RagConfig

rag = RagConfig(
    enabled=True,
    max_documents_length=10000,
    max_retrieved_rag_chunks_count=20,
    max_vector_distance=0.7,
    query_rewrite_prompt_override=(
        "Given the conversation history, rewrite the user's latest query "
        "into a standalone search query that captures the user's intent. "
        "Focus on key entities and concepts."
    ),
)

Using RAG in Agent Prompt Configuration

from elevenlabs.types import PromptAgentApiModelInput, RagConfig

prompt_config = PromptAgentApiModelInput(
    prompt="You are a support agent. Use the knowledge base to answer questions.",
    rag=RagConfig(
        enabled=True,
        max_documents_length=8000,
        max_retrieved_rag_chunks_count=15,
    ),
)

Related Pages

Elevenlabs_Elevenlabs_python_PromptAgentApiModelInput - Parent prompt configuration that references RagConfig
Elevenlabs_Elevenlabs_python_AgentConfig - Core agent configuration
Elevenlabs_Elevenlabs_python_CustomLlm - Custom LLM configuration that may be used alongside RAG

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment