Skip to main content
Rasa helps teams build conversational AI agents that use large language models for natural conversations while keeping business logic explicit and testable.

Choose your starting point

Prototype in your browser

Describe an agent in natural language and test a prototype without installing Rasa.

Build locally

Install Rasa, create a project, and use Rasa Tools in your integrated development environment (IDE).

Understand the Rasa Platform

The Rasa Platform gives developers and conversational AI practitioners two interfaces for working on the same agent. Both interfaces use CALM (Conversational AI with Language Models), Rasa’s approach for using large language models to understand a conversation while flows define the business logic an agent must follow.

What you can build

Explore task-oriented, voice, multilingual, and custom conversational AI use cases.

How the platform works

See how Rasa and Rasa Studio work together across the development lifecycle.

How CALM works

Learn how language understanding and flows combine to keep agents on track.

Choose how your team builds

Build with code, work visually, or use both interfaces on the same agent.

Rasa

Code-firstDefine flows in YAML, connect actions and tools, test behavior, and deploy to infrastructure you control.

Rasa Studio

No-codeDesign and refine flows, manage content, test conversations, and review production interactions in a shared visual workspace.

Continue learning

Build with an AI coding assistant

Use Rasa Tools to inspect, build, and test an agent from your IDE.

Design effective conversations

Plan user journeys and write flows that support natural conversations.