Dev.to AI 🤖 Ai 👁 0 📖 3 min read

Chatbot Development Frameworks: How to Choose

Originally published at heu.ai. What is a chatbot framework, and how is it different from a platform? A framework gives you ready-made parts: handling a conversation, remembering context, calling your systems

Originally published at heu.ai.

What is a chatbot framework, and how is it different from a platform?

A framework gives you ready-made parts: handling a conversation, remembering context, calling your systems and connecting to channels such as a website or a messaging app. You still write the logic. A platform goes further, with a visual builder and hosting, and gives you less control in return.

Which chatbot frameworks are worth looking at?

Three options are worth looking at: Rasa, an open-source conversational AI framework with paid enterprise editions; Botpress, a visual platform for chatbots and agents with a free tier and paid plans; and graph-based frameworks such as LangGraph, for chatbots that use a language model to plan steps and call tools. The Microsoft Bot Framework SDK is deprecated and replaced by the Microsoft 365 Agents SDK, so check older guides that recommend it.

Chatbot frameworks and approaches compared

Option What it is Cost or note
Rasa An open-source framework for building conversational AI. Paid editions for enterprise use.
Botpress A visual platform for building chatbots and AI agents. A free tier and paid plans.
Microsoft Bot Framework SDK Deprecated and replaced by the Microsoft 365 Agents SDK. Check any older guide that recommends it.
Graph-based frameworks An option for chatbots that use a language model to plan steps and call tools. Compared in LangGraph vs CrewAI vs AutoGen.
A model API called directly No framework. The most control and the most to maintain.

How do you choose a chatbot framework?

Ask five questions. How much control do you need, and how fast do you need to launch? Who will maintain it once it is live? Must it run in your own environment, or can it be hosted? Which channels does it need to reach? Does it only answer questions, or must it act on your systems? A framework suits a build that needs control. A platform suits a fast launch with fewer special needs.

When is a framework not enough?

When the chatbot must answer from your own documents, call internal systems, or act on its own, the framework is only one part. You also need retrieval over your documents (RAG development services), integrations with your systems (LLM integration services), and approval steps (agentic AI development services), so a person signs off on any action with real consequence. Assistants like this are covered in our generative AI development services.

How do you build a chatbot on a framework?

Define the job and write down what the chatbot must never do. Choose the approach: framework, platform or a direct model API. Connect your knowledge and your systems. Test with real questions from real users before launch. Deploy with monitoring, and a clear route to a person for anything the chatbot cannot answer.

What should you measure after launch?

Measure three things after launch: whether the chatbot answered the question, how often it handed over to a person, and whether its answers matched your sources. Those three show you where the content or the logic has gaps. Track all three over time, so you can see whether changes to the content or the logic make things better.

Frequently Asked Questions

Is the Microsoft Bot Framework still supported?

The Bot Framework SDK is deprecated and replaced by the Microsoft 365 Agents SDK, which supports C#, JavaScript and Python. Support tickets for the old SDK closed at the end of 2025, so new builds should start on the newer one.

Is an open-source chatbot framework free to run?

The software may be free, but hosting, maintenance and keeping it secure are costs you carry yourself. Paid editions add support and enterprise features.

Can I build a chatbot without a framework?

Yes, by calling a model API directly and writing the conversation handling yourself. It gives the most control, and it leaves you the most to maintain.

Do I need a framework for a chatbot that answers from my documents?

You need retrieval over your documents more than you need any one framework. The framework handles the conversation, and the retrieval layer finds the right passage and shows its source.

Building a chatbot that has to do real work? Book a free consultation and we will look at what the chatbot needs to answer, the systems it has to reach, and the approach that fits.

📰 Read the original article on Dev.to AI

Originally published by Dev.to AI. Aggregated on AIWithGhost for educational purposes — full credit and traffic to the original publisher.