Rasa

Open-source conversational framework that grew CALM: flows you own, LLMs you plug in, and a self-host path for assistants that cannot live only in a vendor canvas.

Product preview

Overview

Rasa predates ChatGPT as the serious open assistant framework and had to add LLMs without throwing away flows: CALM is the bet that conversation still needs structure you can test. Voiceflow is the canvas; Botpress is the studio; Rasa is the framework engineers check into git.

It will not be a Friday widget. It will be the stack when the assistant is a product with evals. Pro packaging and cloud terms have moved — read them. For regulated chat you might self-host, it remains a shortlist item.

Pros / Cons

Pros

Flows you can test, plus an LLM
Self-host path for regulated assistants
Engineer-owned, git-shaped
Older than the toy-bot wave

Cons

Not a CX canvas for Friday
Pro terms have shifted
Setup is a framework, not a paste

Who it's for

Self-hosted assistants with testable flows

Hybrid NLU plus LLM conversation

Replacing a vendor canvas under residency rules

Git-owned dialogue as a product

Bottom line

The engineer-owned assistant framework — pick it when Voiceflow is too closed and a widget is too small.

Key Features

CALM / flow-based assistants

Bring-your-own LLM

Self-host or Rasa Pro

NLU plus LLM hybrid

Open-source core

Compare Rasa with alternatives

Rasa vs Hugging Face Rasa vs Replicate Rasa vs Ollama 4 products · AI Platforms

Frequently asked questions

What is Rasa?

Rasa is a AI Platforms tool developed by Rasa. The company was founded in 2016 and is headquartered in Berlin, Germany. Open-source conversational framework that grew CALM: flows you own, LLMs you plug in, and a self-host path for assistants that cannot live only in a vendor canvas.