Dialogflow vs. Rasa

There are many chatbot platforms with NLP, such as:

• Rasa: Open source conversational AI
• Dialogflow
• Bot Framework
• Wit.ai
• Amazon Lex – Build Conversation Bots

We evaluated the most noteworthy bot platforms for building chatbots for customer support and the service industry. Finally, Dialogflow and Rasa came on top in our priority list.

This article covers the differences between Dialogflow and Rasa and helps you understand which one to choose.

Dialogflow

Dialogflow is a Google-owned product. They have developed human-computer interaction technologies based on natural language conversations.

Dialogflow is a complete closed source product with API and web interface. Dialogflow was previously known as api.ai and was acquired by Google in 2016.

Rasa

Rasa is an open source Machine Learning tool for developers and product teams to expand bots beyond answering simple questions.
Rasa comes with Rasa NLU and Rasa Core. The NLU handles intents and entities while the Core handles dialogues and fulfillment.

Dialogflow vs. Rasa: Major Differences

First of all, we found Dialogflow to be super easy to use. Almost everything is available in the user interface and you can create a bot without writing even a single line of code. Dialogflow provides everything from Entities, Intent, Training, Analytics, History, to Integrations. They are available in the user interface. It is like a full-fledged ready to use service. If you are looking for a Dialogflow alternative, then Rasa is a good option.

In contrast, you will have to build things up on top of Rasa. Also, it requires Python development for customizing the bot.

Projects Providing UI in Rasa

Articulate
Rasa UI

Dialogflow meet everyone’s demand and it is more generalized. In Rasa, you can customize the bot and it requires Python development. Free and paid plans are available in both.

How to Migrate from Dialogflow to Rasa?

Migrating from Dialogflow to Rasa is easy. Dialogflow provides a way to export the agent in zip files. You can then train Rasa NLU using the same downloaded data. Then, modify the API URL to point to Rasa NLU server.

Detailed instructions are available here.

Conclusion

We wanted more control over customization, so we went ahead with Rasa for building our custom bots. For basic bots and customers who are already using Dialogflow, we added support for easy dialogflow integration.

How do you rate Dialogflow vs. Rasa and why? Let us know in the comments below; we would love to hear from you.