OpenAI LLM models | LiveKit docs

Overview

LiveKit Inference includes support for the following OpenAI models. Pricing information for each model and provider is available on the pricing page.

Model name	Model ID	Providers
GPT-4o	openai/gpt-4o	azureopenai
GPT-4o mini	openai/gpt-4o-mini	azureopenai
GPT-4.1	openai/gpt-4.1	azureopenai
GPT-4.1 mini	openai/gpt-4.1-mini	azureopenai
GPT-4.1 nano	openai/gpt-4.1-nano	azureopenai
GPT-5	openai/gpt-5	azureopenai
GPT-5 mini	openai/gpt-5-mini	azureopenai
GPT-5 nano	openai/gpt-5-nano	azureopenai
GPT OSS 120B	openai/gpt-oss-120b	basetengroq(Cerebras coming soon)

Usage

To use OpenAI, pass the model id to the llm argument in your AgentSession. LiveKit Inference manages the connection to the model automatically and picks the best available provider.

from livekit.agents import AgentSession

session = AgentSession(
    llm="openai/gpt-4.1-mini",
    # ... tts, stt, vad, turn_detection, etc.
)

import { AgentSession } from '@livekit/agents';

session = new AgentSession({
    llm: "openai/gpt-4.1-mini",
    // ... tts, stt, vad, turn_detection, etc.
});

Parameters

To customize additional parameters, or specify the exact provider to use, use the LLM class from the inference module.

from livekit.agents import AgentSession, inference

session = AgentSession(
    llm=inference.LLM(
        model="openai/gpt-5-mini", 
        provider="openai",
        extra_kwargs={
            "reasoning_effort": "low"
        }
    ),
    # ... tts, stt, vad, turn_detection, etc.
)

import { AgentSession, inference } from '@livekit/agents';

session = new AgentSession({
    llm: new inference.LLM({ 
        model: "openai/gpt-5-mini", 
        provider: "openai",
        modelOptions: { 
            reasoning_effort: "low" 
        }
    }),
    // ... tts, stt, vad, turn_detection, etc.
});

modelstringRequired

The model to use for the LLM. Must be a model from OpenAI.

providerstringRequired

The provider to use for the LLM. Must be openai to use OpenAI models and other parameters.

extra_kwargsdictOptional

Additional parameters to pass to the provider's Chat Completions API, such as reasoning_effort or max_completion_tokens.

In Node.js this parameter is called modelOptions.

Additional resources

The following links provide more information about OpenAI in LiveKit Inference.

OpenAI Plugin

Plugin to use your own OpenAI account instead of LiveKit Inference.

Azure OpenAI Plugin

Plugin to use your own Azure OpenAI account instead of LiveKit Inference.

OpenAI docs

Official OpenAI platform documentation.

Azure OpenAI docs

Azure OpenAI documentation, for OpenAI proprietary models.

Baseten docs

Baseten's official Model API documentation, for GPT-OSS models.

Groq docs

Groq's official API documentation, for GPT-OSS models.

OpenAI ecosystem overview

Overview of the entire OpenAI ecosystem and LiveKit Agents integration.