Dhtmlx Chatbot

DHTMLX Chatbot is a JavaScript UI widget for building chat interfaces for AI support agents and assistants. It comes from the team behind the DHTMLX JavaScript UI libraries, a long-running set of web components. The widget is released under the MIT licence, so developers can use it in commercial and open-source projects without licence fees. It provides the front end of a chat app, and you connect it to the large language model of your choice.

Features

The widget is lightweight, at around 65KB, and responsive on mobile. It offers four message display modes: blocks, bubbles, cards and flow. A typewriter effect shows replies as they are generated, messages support Markdown, and there is a text area for typing. A sidebar lists chats with different agents, so one interface can host several assistants, and chat history is shown for each conversation. There is also a read-only mode for displaying transcripts.

Connecting a model

DHTMLX Chatbot is model-agnostic. It can be wired to ChatGPT, Claude, Gemini or another LLM through your own backend, which keeps API keys and business logic on the server. The DHTMLX team says it can provide ready-made integration examples and is open to working with other technologies. Typing speed and chat patterns can be configured to suit the product the widget is embedded in.

Documentation and samples

The documentation covers API references and step-by-step guides for initialising, configuring and connecting the widget to a backend. Online samples show its structure, how to place it in a popup and how single and multiple agents work, and you can try a live demo on desktop. The package is delivered by email after you fill in a download form, and support is available through the DHTMLX forum.

Pricing

The widget is free under the MIT licence. You'll still pay for whichever model provider and hosting you use behind it.

Things to consider

DHTMLX Chatbot is a UI component, not a complete chatbot service. You have to build or supply the backend that calls the model, manages conversations and enforces any safety or data rules. It suits developers who want a polished chat front end that they control, rather than teams looking for a hosted, no-code bot builder.