Struggling to get Rasa running on your desktop?

Concerned about providing access to your desktop for user testing?

Connect to a hosted Rasa instance in seconds with our Rasa-as-a-Service.

It’s easy (and free) to get started.

yellow-line provides an alternative environment to run and test your Rasa models in AWS easily and securely – without any of the configuration issues from running it locally and the security concerns of providing access to end-users for testing.

Sign up for BETA program

By clicking “Sign up for”, you agree to our Terms of Service and Privacy Statement.

What is Rasa?

Rasa is an open source machine learning framework to automate text- and voice-based conversations – developers can access tools to build, improve, and deploy text-and voice-based chatbots and assistants.


What is provides a SaaS environment to quickly spin up Rasa instances and run your models, making it easier and more secure for end-user testing, easier to evaluate configurations and pipelines, and simple to check version compatibility – it’s Rasa-as-a-Service.

# Install with NPM
npm i @raas-ai/raas-cli



Instant Rasa to run your model


Command line access from local desktop


Faster training of your model


Automatic identification of best pipeline configuration


Diagnostic information and performance tracking


Add & customize messenger for end-user testing


Enables A/B Testing


Compatibility testing against new Rasa versions


Maintain and store multiple versions


No need to install Rasa locally provides an instant Rasa instance for you to run your model.

Really easy-to-use command line

We take the pain out of configuration and python library conflicts and make it really easy to use from your desktop.

Secure end-user access for testing

We allow you to provide safe and secure access for end-users to test your models without exposing your desktop (users can test without your desktop being on).

Overcome Mac/Win desktop issues

We overcome the problems of running Rasa on Windows and Mac machines.

Generate higher accuracy models

We help you get to higher levels of accuracy and F1 scores much faster than natively

Track & evidence improvements over time

Monitor improvements and show team members your progress over time.

Easy to compare and maintain versions

Let us take the pain of maintaining and storing versions as well as compatibility checks with new releases of Rasa

Instant deployment to your website

When you’re ready to use, simply add your customized messenger onto your website with a few lines of Javascript.


Startup Time
Concurrent Models
Production Models
Model Versions
A/B Live Testing
Chat Messenger
Chat History
Security & Access Logs
Action Server
Github Integration


5 - Projects
1 - Users
Instant - Startup Time
2 - Concurrent Models
0 - Production Models
Limited - Library
Last 10 - Model Versions
raas-cli - Analytics
No - A/B Live Testing
No - Chat Messenger
1 week - Chat History
No - Security & Access Logs
No - Action Server
No - Github Integration


15th January 2020

Rasa comparison with Google Dialogflow and Amazon Lex AI Engines

Building natural language AI assistants is a hot topic in the business and developer worlds so let’s…

15th January 2020

An Introduction to, a tool to train, host and analyze AI models.

Here’s a short video introducing, a tool to train, host and analyze AI models, Creating a…

15th January 2020

How makes it easier to build and scale solutions based on Rasa

The Rasa Open-Source Framework In our last blog, we looked at the popularity of publicly available NLU…