Product

Agent Builder

Build and test production-ready voice agents. Configure your agent in a few simple steps, preview it live, then deploy it to LiveKit Cloud with no local setup required.

Start building

Documentation

Illustration of isometric grid with Agent Builder elements

Deploy in less than 5 minutes

1-click deploy to LiveKit Cloud

300+ AI model integrations

Prototype voice agents in your browser

Quickly build a proof-of-concept, explore ideas, or stand up a working voice agent in minutes. Every agent built on Agent Builder is fully executable Python code.

Customize prompt
Choose model
Define actions
Preview and deploy
Convert to code

Write instructions to define your voice agent’s identity and behavior. Give it a name and a welcome greeting when users first join the call.

Prototype voice agents in your browser

Quickly build a proof-of-concept, explore ideas, or stand up a working voice agent in minutes. Every agent built on Agent Builder is fully executable Python code.

Customize prompt

Write instructions to define your voice agent’s identity and behavior. Give it a name and a welcome greeting when users first join the call.

Choose model

Shape your agent’s voice, character, and language by selecting AI models. Use a cascaded STT-LLM-TTS pipeline or try xAI’s Grok realtime model.

Define actions

Give your agent the ability to interact with web-based APIs, retrieve data, connect to MCP servers, and define data collection sequences with configurable tools.

Preview and deploy

Preview your agent by starting a conversation in your browser. When you like what you hear, deploy it to production with a single click.

Convert to code

Download your agent code from the LiveKit Cloud dashboard and continue to iterate in your IDE with our Python Agents SDK.

The LiveKit Platform

Build, run, and observe
agents with LiveKit Cloud

Our end-to-end platform powers enterprise-grade voice AI
for customer support at global scale.

Explore the full platform

Resources to help you learn and ship

Watch a step-by-step tutorial video

video

Learn why we built Agent Builder

product

Build a voice AI frontend in less than 10 min

guide

FAQs

How do I get started with LiveKit’s Agent Builder?

Sign up for a LiveKit Cloud account, then select “Deploy a new agent” in your project’s Agents dashboard to access Agent Builder.

How does Agent Builder compare to other low-code / no-code voice AI tools?

Voice agents built with Agent Builder are complete, production-ready agent backends written in Python. Agent Builder is designed as a pathway to code, so you can get started quickly in your browser, then continue to iterate in your IDE. Other low-code / no-code voice AI tools are configuration-based rather than code-based.

Every agent deployed using Agent Builder is a real, fully functional LiveKit Agent running on LiveKit Cloud and is no different than an agent deployed from code. Unlike other low-code / no-code voice AI platforms, agents deployed on LiveKit Cloud do not use a shared runtime with other users and are not subject to automated version updates.

What voice AI models are available in Agent Builder?

Agent Builder supports all models available with LiveKit Inference, our unified API for calling STT, LLM, TTS, and realtime models. For a full list of available models, visit our pricing page.

When should I use Agent Builder vs the Agents SDKs?

Use Agent Builder to quickly prototype and iterate on basic voice agents without leaving your browser. If you prefer to start with code or use a coding assistant, use the Agents SDKin your preferred language. Any agent built using Agent Builder can be exported to Python code and edited later in your IDE.

Can I convert agents into code later?

Yes, you can download the complete code for your agent directly from Agent Builder to keep iterating with our Python Agents SDK. To learn more, visit our docs.

How do I modify the code of my agent?

To modify your agent code, download it from Agent Builder. You can then import the files into your code editor to keep iterating on your voice agent with our Python Agents SDK. To learn more, visit our docs.

How do I add tools or custom logic to agents?

Agent Builder offers a few different options to give your agents the ability to take action by accessing external systems and services, including HTTP tools, client tools, and MCP servers. To learn more, visit our docs.

How do I deploy agents on Agent Builder?

You can deploy agents to production with a single click in Agent Builder. Your agent will be deployed to LiveKit Cloud and ready to integrate with a custom frontend or phone number. To learn more, visit our docs.

Ready to build?

Start building a voice AI agent with a free account. Reach out to us if you're interested in custom pricing.

Start building

Contact sales

No credit card required • 1,000 free agent session minutes monthly

Agent Builder