AI-Enabled Messaging Experience
Picture a chat that effortlessly starts and flows, like catching up with your best buddy over a cup of coffee. With CometChat AI, we’re making that a reality! Introducing Conversation Starter and Smart Replies that ignite natural and organic conversations. Say goodbye to awkward silences and hello to chat-filled adventures! For those who require a concise overview of their discussions, the Conversation Summary feature is available to provide a succinct recap. These features are powered by OpenAI. You configure your OpenAI model and API key once in the CometChat dashboard (below), enable each feature, and then either let the UI Kits render them automatically or fetch them yourself through the SDK.Conversation Starter
AI-generated opening lines for a new conversation.
Smart Replies
Contextual positive, negative, and neutral reply suggestions.
Conversation Summary
A short recap of a long or unread conversation.
Pre-requisite
- Login to your CometChat dashboard and choose your app.
- Navigate to AI Chatbot > Settings in the left-hand menu.
Set the GPT Model
Enter the name of the OpenAI model that you intend to use.Enter the exact OpenAI model name as it appears in OpenAI’s documentation (for example, the value you would pass as the
model parameter to the OpenAI API). For the current list of supported model names, refer to OpenAI’s models documentation. If a model name is not accepted, confirm it is available on your OpenAI account.Save the OpenAI Key
You can get the OpenAI Key from your OpenAI account. This will be used by CometChat to interact with the OpenAI APIs.Set a Custom Instruction
Custom Instruction is an information which gets added in each and every ChatGPT prompt made by the CometChat AI. Custom Instruction is app-level information you can add to describe your use-case & inform what kind of responses you need from the CometChat AI.Set the Temperature
The API is non-deterministic by default. This means that you might get a slightly different completion every time you call it, even if your prompt stays the same. Setting temperature to 0 will make the outputs mostly deterministic, but a small amount of variability will remain. Lower values for temperature result in more consistent outputs, while higher values generate more diverse and creative results. Select a temperature value based on the desired trade-off between coherence and creativity for your specific application.Enable AI
Toggle on AI.Rendering AI features in your app
After the settings above are saved and AI is enabled, turn on the individual features you want (Conversation Starter, Smart Replies, Conversation Summary) and surface them in your app in one of two ways:- UI Kits (no extra code): Enable each feature from the dashboard and the UI Kit renders it automatically — Conversation Starter appears in the message list, while Smart Replies and Conversation Summary appear in the composer’s action sheet. See Smart Chat Features in the React UI Kit for exactly which component shows each feature.
- SDK methods (full control): Fetch suggestions yourself and render your own UI. Each feature page documents its SDK method and configuration — see Conversation Starter, Smart Replies, and Conversation Summary. For the JavaScript SDK reference, see the AI User Copilot SDK overview.