Ferhat Gölge
Computer Worker
What is the AI SDK that comes with Laravel 13? How to use it?
One of the most important new features in Laravel 13 is built-in AI SDK support. Laravel developers can now develop AI features directly within the framework, without needing extra packages for integration with services like OpenAI, Gemini, and Anthropic.
In this article, you will learn what Laravel 13 AI SDK is, how to use it, what can be done with it, and how it works with a simple example.
🧠 What is Laravel AI SDK?
Laravel AI SDK is a layer that brings together different AI providers under a single API.
So think about this:
You previously wrote separate code for OpenAI and for Gemini.
👉 Now you manage them all with a single Laravel API.
Supported providers:
- OpenAI
- Google Gemini
- Anthropic (Claude)
- Mistral
- Groq
- Ollama (local AI)
- Azure OpenAI
⚡ What can you do with the Laravel AI SDK?
With Laravel 13, you can easily develop the following AI scenarios:
🎯 Text Processing
- Creating a blog post
- Generating product descriptions
- Text summarization
- Language translation
🤖 Chatbot & Agent
- Customer support bot
- CV analysis system
- AI consultant
🧾 Structured Output
- Generating data in JSON format.
- Automatic form filling
🔍 Embedding & Search
- Semantic search (RAG system)
- AI-powered filtering
🎙️ Audio & Visuals
- Speech to text
- Text to speech
- Image generation
🏗️ Installation
First, install the package:
Next, config and migration:
php artisan migrate
Add the API key to your .env file:
⚙️ System Requirements
For Laravel 13 AI SDK:
- PHP 8.3+
- Laravel 13
- Composer
- API Key (OpenAI, etc.)
Additional important point:
👉 No GPU required (if you're using cloud AI)
👉 If you use the local model (Ollama) → RAM/CPU will be required.
🧩 Basic Logic: Agent Structure
Laravel AI SDK goes a step further than the classic "send prompt" logic.
Main structure:
- Agent → AI character
- Instruction → tells you what to do.
- Prompt → user request
- Memory → speech history
🧪 Simple Example: Creating a Product Description
1. Create an Agent
use Laravel\Ai\Contracts\Agent;
use Laravel\Ai\Promptable;
class ProductDescriptionAgent implements Agent
{
use Promptable;
public function instructions(): string
{
return 'You are a professional e-commerce copywriter. Write short product descriptions in Turkish.';
}
}
2. Usage
Route::get('/ai-demo', function () {
$response = ProductDescriptionAgent::make()->prompt(
'Write a short description for a black leather men's wallet'
);
return (string) $response;
});
🎉 It's out!
📦 Structured Output (JSON Output)
You can get not only text but also structured data from AI:
For example:
"title": "Product name"
"description": "Description"
Rating: 4.5
}
This feature is especially important for:
- Admin panel
- Automation systems
- API outputs
It is very powerful.
🔥 Streaming (Live Typing)
With the Laravel AI SDK, you can create typing display features similar to ChatGPT:
In this way:
- Chat application
- Live support
- AI editor
Systems like these can be created.
🧠 Embedding & AI Search
One of the most powerful features that comes with Laravel 13:
👉 Semantic search (AI search)
Example:
- User: "cheap shoes"
- System: finds the products that are closest in meaning.
This structure:
- Blog search
- Product filtering
- SaaS platforms
It is very critical.
⚖️ Advantages
The biggest advantages of Laravel AI SDK:
✔ Multiple AI support with a single API
✔ Clean and sustainable code structure
✔ Agent architecture
✔ Structured output
✔ Streaming support
✔ RAG / embedding infrastructure
⚠️ Important Considerations
- AI services are paid (token-based).
- Rate limit management should be implemented.
- Using caching and queries is recommended.
- Sensitive data should be sent with care.
🎯 Real-Life Use Cases
Here are a few examples specific to your projects:
- Blog post generator in the admin panel.
- CV builder (adding AI to KitResume)
- Automatic classification of customer messages
- Product description generation (e-commerce)
- AI-powered search system
⚔️ Laravel 13 AI SDK vs. Old AI Integration Method
| Feature | Laravel 13 AI SDK | Old Method (Laravel 10-12 + Packages) |
|---|---|---|
| Setup | Single package ( laravel/ai ) |
Separate packages for each service. |
| API Structure | Unified (single API) | Disorganized, dependent on the provider. |
| Changing Provider | Very easy (with config) | You need to change the code. |
| Agent Structure | Native support | Written manually. |
| Prompt Management | Central (Agent class) | Distributed within the controller |
| Structured Output | Built-in | Manual parsing |
| Streaming | Native ( ->stream() ) |
Additional package required. |
| Conversation Memory | Ready-made (DB-based) | You need to write it yourself. |
| Tool Calling | Built-in | Complicated |
| Embedding | Native support | External service + manual |
| Vector Search | Integrated | Additional infrastructure is required. |
| Local AI (Ollama) | Supported | Additional integration is required. |
| Code Cleanup | Very high | Middle |
| Maintainability | High | Low / Medium |
| Learning Curve | Easier | More difficult |
| Development Rate | Very fast 🚀 | Slow |
📌 Result
With Laravel 13, AI integration is now available:
👉 no extra package required
👉 native
👉 scalable
It transformed into a structure.
This makes Laravel more than just a backend framework,
It is transforming into an AI-powered application development platform .
Keywords: Laravel 13 AI, what is Laravel AI SDK, Laravel OpenAI integration, Laravel AI agent usage
Frequently Asked Questions
What is Laravel AI SDK?
Laravel AI SDK is a built-in integration layer that allows developers to interact with multiple AI providers using a single API.
Does Laravel AI require GPU?
No, if you use cloud-based AI providers like OpenAI, no GPU is required.
Can Laravel AI work with local models?
Yes, Laravel supports local AI models via tools like Ollama. bunların bir de Türkçelerini verir misin?