Implementation:Elevenlabs Elevenlabs python ConversationalAIRawClient

Field	Value
Sources	src/elevenlabs/conversational_ai/raw_client.py, src/elevenlabs/conversational_ai/client.py
Domains	Conversational AI, Knowledge Base, RAG Indexing
Last Updated	2026-02-15

Overview

The RawConversationalAiClient provides the low-level HTTP implementation for the Conversational AI namespace in the ElevenLabs API. It handles knowledge base document management and RAG (Retrieval-Augmented Generation) index operations. Unlike the high-level ConversationalAiClient which returns parsed model objects, the raw client returns HttpResponse wrapper objects that include both the parsed data and the full HTTP response metadata (status code, headers, etc.).

The raw client is accessible via client.conversational_ai.with_raw_response. The high-level ConversationalAiClient also serves as a namespace that provides lazy access to 17 sub-clients: conversations, twilio, whatsapp, agents, tests, phone_numbers, llm_usage, knowledge_base, tools, settings, secrets, batch_calls, sip_trunk, mcp_servers, whatsapp_accounts, analytics, and dashboard.

Both synchronous (RawConversationalAiClient) and asynchronous (AsyncRawConversationalAiClient) variants are provided.

API Endpoints

Method	HTTP	Endpoint
add_to_knowledge_base()	POST	v1/convai/knowledge-base
rag_index_overview()	GET	v1/convai/knowledge-base/rag-index
get_document_rag_indexes()	GET	v1/convai/knowledge-base/{documentation_id}/rag-index
delete_document_rag_index()	DELETE	v1/convai/knowledge-base/{documentation_id}/rag-index/{rag_index_id}

Code Reference

Source Location

• Raw client: src/elevenlabs/conversational_ai/raw_client.py
• High-level client: src/elevenlabs/conversational_ai/client.py

Class Signatures

class RawConversationalAiClient:
    def __init__(self, *, client_wrapper: SyncClientWrapper): ...

class AsyncRawConversationalAiClient:
    def __init__(self, *, client_wrapper: AsyncClientWrapper): ...

Import

The raw client is accessed through the high-level client's with_raw_response property:

from elevenlabs import ElevenLabs

client = ElevenLabs(api_key="YOUR_API_KEY")
raw_client = client.conversational_ai.with_raw_response  # RawConversationalAiClient instance

I/O Contract

add_to_knowledge_base()

Upload a file or webpage URL to create a knowledge base document. After creating the document, update the agent's knowledge base by calling the Update agent endpoint.

Input Parameters:

Parameter	Type	Required	Description
agent_id	typing.Optional[str]	No	The agent ID to associate this document with (passed as query parameter)
name	typing.Optional[str]	No	A custom, human-readable name for the document
url	typing.Optional[str]	No	URL to a page of documentation that the agent will have access to
file	typing.Optional[core.File]	No	File to upload as a knowledge base document
request_options	typing.Optional[RequestOptions]	No	Request-specific configuration

Output:

Type (Raw)	Type (High-level)	Description
HttpResponse[AddKnowledgeBaseResponseModel]	AddKnowledgeBaseResponseModel	Created knowledge base document details

rag_index_overview()

Provides total size and other information of RAG indexes used by knowledgebase documents.

Input Parameters:

Parameter	Type	Required	Description
request_options	typing.Optional[RequestOptions]	No	Request-specific configuration

Output:

Type (Raw)	Type (High-level)	Description
HttpResponse[RagIndexOverviewResponseModel]	RagIndexOverviewResponseModel	Overview of RAG index usage and total sizes

get_document_rag_indexes()

Provides information about all RAG indexes of the specified knowledgebase document.

Input Parameters:

Parameter	Type	Required	Description
documentation_id	str	Yes	The id of a document from the knowledge base. Returned on document addition.
request_options	typing.Optional[RequestOptions]	No	Request-specific configuration

Output:

Type (Raw)	Type (High-level)	Description
HttpResponse[RagDocumentIndexesResponseModel]	RagDocumentIndexesResponseModel	All RAG indexes for the specified document

delete_document_rag_index()

Delete a RAG index for a knowledgebase document.

Input Parameters:

Parameter	Type	Required	Description
documentation_id	str	Yes	The id of a document from the knowledge base. Returned on document addition.
rag_index_id	str	Yes	The id of the RAG index of the document from the knowledge base.
request_options	typing.Optional[RequestOptions]	No	Request-specific configuration

Output:

Type (Raw)	Type (High-level)	Description
HttpResponse[RagDocumentIndexResponseModel]	RagDocumentIndexResponseModel	Deleted RAG index details

Error Handling

All methods may raise:

• UnprocessableEntityError (HTTP 422) -- contains an HttpValidationError body with validation details.
• ApiError -- for all other non-2xx HTTP status codes.

Sub-Client Properties (High-Level Client)

The ConversationalAiClient provides lazy access to the following sub-clients:

Property	Description
conversations	Manage conversations
twilio	Twilio integration
whatsapp	WhatsApp integration
agents	Manage conversational AI agents
tests	Test conversational AI agents
phone_numbers	Manage phone numbers
llm_usage	LLM usage tracking
knowledge_base	Knowledge base management
tools	Manage agent tools
settings	Conversational AI settings
secrets	Manage secrets
batch_calls	Batch call operations
sip_trunk	SIP trunk configuration
mcp_servers	MCP server management
whatsapp_accounts	WhatsApp account management
analytics	Analytics and reporting
dashboard	Dashboard data access

Usage Examples

Add a Document to Knowledge Base (High-Level)

from elevenlabs import ElevenLabs

client = ElevenLabs(
    api_key="YOUR_API_KEY",
)
response = client.conversational_ai.add_to_knowledge_base(
    agent_id="agent_id",
)

Add a Document via URL

from elevenlabs import ElevenLabs

client = ElevenLabs(
    api_key="YOUR_API_KEY",
)
response = client.conversational_ai.add_to_knowledge_base(
    agent_id="agent_id",
    name="Product Documentation",
    url="https://docs.example.com/product-guide",
)

Get RAG Index Overview

from elevenlabs import ElevenLabs

client = ElevenLabs(
    api_key="YOUR_API_KEY",
)
overview = client.conversational_ai.rag_index_overview()

Get Document RAG Indexes

from elevenlabs import ElevenLabs

client = ElevenLabs(
    api_key="YOUR_API_KEY",
)
indexes = client.conversational_ai.get_document_rag_indexes(
    documentation_id="21m00Tcm4TlvDq8ikWAM",
)

Delete a Document RAG Index

from elevenlabs import ElevenLabs

client = ElevenLabs(
    api_key="YOUR_API_KEY",
)
result = client.conversational_ai.delete_document_rag_index(
    documentation_id="21m00Tcm4TlvDq8ikWAM",
    rag_index_id="21m00Tcm4TlvDq8ikWAM",
)

Raw Response Access

from elevenlabs import ElevenLabs

client = ElevenLabs(api_key="YOUR_API_KEY")

# Use the raw client to get full HTTP response metadata
raw_response = client.conversational_ai.with_raw_response.rag_index_overview()
print(raw_response.response.status_code)
print(raw_response.response.headers)
print(raw_response.data)

Async Usage

import asyncio
from elevenlabs import AsyncElevenLabs

client = AsyncElevenLabs(
    api_key="YOUR_API_KEY",
)

async def main() -> None:
    # Add document to knowledge base
    response = await client.conversational_ai.add_to_knowledge_base(
        agent_id="agent_id",
    )

    # Get RAG index overview
    overview = await client.conversational_ai.rag_index_overview()

    # Get document RAG indexes
    indexes = await client.conversational_ai.get_document_rag_indexes(
        documentation_id="21m00Tcm4TlvDq8ikWAM",
    )

    # Delete a document RAG index
    result = await client.conversational_ai.delete_document_rag_index(
        documentation_id="21m00Tcm4TlvDq8ikWAM",
        rag_index_id="21m00Tcm4TlvDq8ikWAM",
    )

asyncio.run(main())

Related Pages

• Elevenlabs_Elevenlabs_python_WebhooksClient -- Webhook management for workspace events
• Elevenlabs_Elevenlabs_python_TextToVoiceRawClient -- Voice design and generation from text descriptions
• Elevenlabs_Elevenlabs_python_Environment -- Environment and base URL configuration