AI features for your app

1. 1

   Decide what AI feature you want to add to your app, such as a chatbot, image generator, or text summarizer.

   Lovable's built-in AI connector can implement many different AI features, so knowing your goal helps you ask for the right thing.

2. 2

   If you want to manage how Lovable uses AI, go to 'Connectors', select 'AI', and adjust settings under 'Manage my agent's permissions'.

   This lets you control whether Lovable automatically adds AI features, asks for your approval each time, or never allows it.

3. 3

   Ask Lovable to add the desired AI feature to your app, describing what you want to build.

   Lovable will set up the necessary AI infrastructure, including API keys and secure backend calls, so you don't have to.

4. 4

   If you have specific needs for speed, quality, or model type, mention them in your request.

   Lovable can pick the best model for your use case (e.g., 'fast' for quick responses, 'highest quality' for images, or a specific model name if you know it).

Add a chatbot to my app that can answer questions about my product documentation. It should be able to summarize long text and understand user feedback.

API key

A secret code that allows your app to talk to an online service, like an AI provider, proving it's allowed to use that service.

Backend

The part of your app that runs on a server and handles things users don't directly see, like processing data or making secure calls to other services.

Streaming

When an AI's response appears piece by piece, like words typing out one at a time, instead of waiting for the whole answer to be ready before showing it.

Semantic search

Searching for information based on the meaning of your words, rather than just matching exact keywords.

Retrieval-augmented generation (RAG)

An AI technique where the AI first looks up relevant information from a knowledge base and then uses that information to generate a more accurate and informed answer.

Zero-data-retention models

AI models that do not store any of the data you send to them, which is important for privacy and security.

Edge function

A small piece of code that runs very close to where your users are, making your app faster and more responsive.

Server-sent events (SSE)

A way for a web server to automatically send new information to your app as it becomes available, without your app having to constantly ask for updates.

• Don't try to manually set up API keys or billing for AI providers; Lovable handles this for you.
• Be aware that showing an AI model's 'thinking' or using 'priority processing' can increase costs and response times.
• Remember that video generation takes longer (1-2 minutes) and costs more than chat or image generation, and generated clips are only available for 48 hours unless you save them.
• If you need predictable costs and responses for chat, choose a numbered GPT model instead of 'Chat Latest', as the underlying model and its pricing can change.