Direct answer
What is Rasa?
Rasa is a conversational AI platform designed to build, orchestrate, and govern enterprise chatbots and voice agents. Using its CALM framework, it merges generative LLMs with deterministic logic, allowing teams to develop via pro-code frameworks or no-code visual builders and deploy on self-hosted or private infrastructure.
Rasa is an enterprise-grade conversational AI platform built to deliver controllable, production-ready AI agents across text and voice channels. Rather than relying purely on prompt-based models that risk hallucinations, Rasa uses CALM (Conversational AI with Language Models), a dialogue management system that combines the linguistic flexibility of LLMs with structured business logic, deterministic flows, and built-in error recovery patterns. The platform is divided into pro-code and no-code environments. Rasa Pro provides software engineers with an extensible Python framework, command-line tooling, SDKs for custom action servers, and Model Context Protocol (MCP) tool integrations. Rasa Studio provides conversational designers and business teams with a visual interface for managing flows, authoring channel- and language-specific responses, testing interactions, and reviewing live conversation histories. Designed for highly regulated sectors such as banking, healthcare, insurance, and government, Rasa supports flexible deployment topologies. Teams can run agents entirely on their own infrastructure—including private clouds, Kubernetes clusters via Helm charts, and air-gapped environments—or use managed services. It also integrates real-time voice and IVR capabilities with low-latency turn-taking, barge-in support, and streaming speech handling.
Unlike pure LLM wrapper frameworks or rigid intent trees, Rasa's CALM architecture anchors generative language models inside explicit, deterministic business flows and supports air-gapped, on-premises deployment.
Product capabilities
Key features
CALM Dialogue Management
Combines language model understanding with deterministic state handling, guided workflows, and recovery patterns to avoid black-box failures.
Rasa Studio No-Code Builder
A collaborative visual interface enabling business teams to design flows, manage multi-channel content, edit prompts, and review conversation logs.
Model Context Protocol (MCP) Integration
Standardizes tool execution by allowing agents to connect to external systems, databases, and trade or inventory servers as declared MCP tools.
On-Premises and Air-Gapped Deployment
Supports deployment on private cloud, local clusters via Kubernetes Helm charts, or offline air-gapped environments for complete data privacy.
Real-Time Voice and IVR
Low-latency voice infrastructure featuring streaming speech recognition, turn-taking, barge-in management, and telephony connectors like AudioCodes.
Multi-Agent Orchestration
Coordinates networks of specialized subagents with shared memory and state handoffs to handle complex, multi-domain user workflows.
Enterprise Observability & Security
Includes OpenTelemetry observability, PII redaction, HashiCorp Vault secrets management, Redis multi-node concurrency, and Kafka event pipelines.
Workflow
How Rasa works
- Define conversational logic, intents, and business actions using the pro-code developer framework or the Rasa Studio visual interface.
- Connect backend APIs, enterprise RAG knowledge bases, and tools using the Rasa SDK or the Model Context Protocol (MCP).
- Deploy the assistant on your chosen infrastructure, such as self-hosted Kubernetes or private cloud environments.
- Run and monitor conversations with real-time observability, evaluating decision paths and optimizing responses across chat and voice channels.
Practical fit
Who should use Rasa?
Banking and Financial Service Automation
Automating account transactions, fraud monitoring, loan qualification workflows, and portfolio inquiries under strict regulatory compliance.
Healthcare Navigation and Scheduling
Deploying patient-facing assistants that triage symptoms, book provider appointments, and answer policy questions while protecting health data.
Omnichannel Customer Support
Deflecting routine customer service tickets across web, mobile apps, WhatsApp, and telephony with automated self-service and graceful handoffs.
Internal IT and HR Helpdesks
Orchestrating internal enterprise requests, employee onboarding, policy inquiries, and backend ticketing integrations.
Editorial assessment
Pros and limitations
Where it is strong
- Supports air-gapped, on-premises, and private cloud deployment to meet strict data privacy standards
- Combines pro-code Python SDK flexibility with Rasa Studio visual flow editing
- CALM architecture prevents uncontrolled model hallucination across multi-step transactional flows
- Generous free Developer Edition available for local or production use up to 1,000 external sessions per month
- Native support for Model Context Protocol (MCP) and real-time voice primitives
Where to be careful
- Steep learning curve compared to lightweight no-code widget builders
- Enterprise features such as Studio, SSO, RBAC, and telephony integrations require contacting sales
Commercial context
Rasa pricing
At the review date (September 2026), Rasa offers a free Developer Edition for up to 1,000 external or 100 internal conversations per month. Enterprise tiers (Rasa Pro and Rasa Studio) require contacting sales.
| Plan | Price | What it includes |
|---|---|---|
| Developer Edition | $0 / month (free license for local or production use; 1 bot per company, up to 1,000 external or 100 internal conversations/month with community forum support) | |
| Enterprise (Rasa Pro / Rasa Studio) | Custom / annual or contact sales (includes full Rasa Platform, pro-code Pro framework, no-code Studio, RBAC, SSO, Helm deployment, and premium support) |
Pricing, limits, taxes, model access, and regional availability can change. Verify the purchase-critical details on the official pricing page linked under Sources.
Transparent ranking
Why Rasa scores 80.1
Each factor is scored on a 100-point scale, then combined using the public ToolsRank weights. Engagement and momentum stay at a neutral baseline until measured signals exist, so no tool can gain or lose position from numbers nobody recorded.
Compatibility
Languages, platforms, and integrations
Languages
- Multilingual support with language-agnostic NLU
Platforms
- Self-Hosted
- Kubernetes
- Linux
- Docker
- Web
Integrations & surfaces
- Facebook Messenger
- Salesforce
- AudioCodes VoiceAI Connect
- Kafka
- Redis
- HashiCorp Vault
- OpenTelemetry
- Kubernetes (Helm)
- Cursor
- Claude Code
- VS Code
Community
Reviews and questions
No approved member reviews yet. Editorial factors above are the only rating on this page.
Reviews and questions come from Google-signed members and are checked by an editor before they appear.
Frequently asked
Rasa FAQ
What is Rasa CALM?+
CALM stands for Conversational AI with Language Models. It is Rasa's dialogue management architecture that uses LLMs to interpret user language while steering interactions through deterministic, business-governed workflows to avoid unexpected model behaviors.
Is Rasa open source or free to use?+
Rasa provides a free Developer Edition that can be used locally or in production for one bot per company, handling up to 1,000 external or 100 internal conversations per month with community forum support. Larger scale deployments require an Enterprise plan.
Can Rasa be deployed on-premises or in an air-gapped network?+
Yes. Rasa supports self-managed deployments across bare metal, private clouds, Kubernetes clusters via Helm, and fully offline or air-gapped environments, ensuring that sensitive data never leaves your infrastructure.
What is the difference between Rasa Pro and Rasa Studio?+
Rasa Pro is the pro-code generative AI framework for developers, offering SDKs, CLI tools, testing, and deployment configurations. Rasa Studio is a no-code visual interface layered on top of Rasa Pro that lets conversation designers and business users build flows, edit responses, and review conversation logs.
