Implementation:Elevenlabs Elevenlabs python CustomLlm

Field	Value
source	Elevenlabs_Elevenlabs_python
domains	Conversational AI, Custom LLM, API Integration
last_updated	2026-02-15

Overview

Description

CustomLlm is a Pydantic model that defines the configuration for integrating a custom LLM endpoint with an ElevenLabs conversational AI agent. It allows developers to use their own Chat Completions-compatible endpoint instead of the built-in LLM options, providing flexibility to use self-hosted or third-party language models.

This model is auto-generated by Fern from the ElevenLabs API definition and inherits from UncheckedBaseModel. It supports authentication via API keys, custom request headers, API versioning, and multiple API types (chat completions or responses).

Usage

CustomLlm is used within PromptAgentApiModelInput when the agent's LLM is set to CUSTOM_LLM. It enables the agent to route LLM requests to a user-specified endpoint, such as an Azure OpenAI deployment, a self-hosted model, or any Chat Completions-compatible service.

Code Reference

Source Location

src/elevenlabs/types/custom_llm.py

Class Signature

class CustomLlm(UncheckedBaseModel):
    ...

Import Statement

from elevenlabs.types import CustomLlm

I/O Contract

Field	Type	Required	Description
`url`	`str`	Yes	The URL of the Chat Completions compatible endpoint.
`model_id`	`Optional[str]`	No	The model ID to be used if URL serves multiple models.
`api_key`	`Optional[ConvAiSecretLocator]`	No	The API key for authentication.
`request_headers`	`Optional[Dict[str, CustomLlmRequestHeadersValue]]`	No	Headers that should be included in the request.
`api_version`	`Optional[str]`	No	The API version to use for the request.
`api_type`	`Optional[CustomLlmapiType]`	No	The API type to use (chat_completions or responses).

Usage Examples

Basic Custom LLM Configuration

from elevenlabs.types import CustomLlm

custom_llm = CustomLlm(
    url="https://my-llm-service.example.com/v1/chat/completions",
    model_id="my-fine-tuned-model",
)

Custom LLM with Azure OpenAI

from elevenlabs.types import CustomLlm

custom_llm = CustomLlm(
    url="https://my-deployment.openai.azure.com/openai/deployments/gpt-4/chat/completions",
    api_version="2024-02-15-preview",
)

Using Custom LLM in Agent Prompt Config

from elevenlabs.types import PromptAgentApiModelInput, CustomLlm

prompt_config = PromptAgentApiModelInput(
    prompt="You are a helpful assistant powered by a custom model.",
    custom_llm=CustomLlm(
        url="https://api.example.com/v1/chat/completions",
        model_id="custom-model-v2",
    ),
)

Related Pages

Elevenlabs_Elevenlabs_python_PromptAgentApiModelInput - Parent prompt configuration that references CustomLlm
Elevenlabs_Elevenlabs_python_AgentConfig - Core agent configuration
Elevenlabs_Elevenlabs_python_GetAgentResponseModel - Full agent response model

Page Connections

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

Principle

Implementation

Heuristic

Environment