Expose a local LLM API on your LAN safely
Use a local OpenAI-compatible endpoint without accidentally exposing it to the public Internet
Read tutorial →AI & Local LLMs
Build with AI. Run it locally.
Learn how local language models work, choose models that fit your hardware, build useful AI workflows, and connect AI to real web applications.
Start here
What this section will teach you.
The goal here is practical AI: understanding models, running them on your own hardware, exposing them safely to applications, and deciding when local AI is a better fit than a hosted API.
01
Understand tokens, context windows, quantization, and model sizes
02
Choose models that fit available RAM and VRAM
03
Install and configure tools such as LM Studio
04
Run local inference and test prompts systematically
05
Expose a local model through an API on your LAN
06
Connect PHP, Python, or JavaScript applications to a model
07
Design reliable prompts and structured outputs
08
Measure speed, quality, memory use, and tradeoffs
Learning path
Follow the ideas in order.
You can jump directly to the subject you need, or work through the path from left to right.
01AI FoundationsOpen lesson → 02Pick a ModelOpen lesson → 03Run LocallyOpen lesson → 04Prompt DesignOpen lesson → 05Local APIsOpen lesson →
06Web Integration
07Automation
08Performance Tuning
Published guides
Tutorials & projects
New articles assigned to this category appear here automatically.
Connect a PHP website to a local LLM
Send structured requests from PHP and handle model responses safely
Read tutorial →AI foundations: tokens, context windows, and quantization
Understand the practical vocabulary you need before choosing a local model
Read tutorial →Choose a local LLM that fits your RAM and VRAM
Match model size and quantization to the hardware you actually own
Read tutorial →Run a model locally with LM Studio
Install a local inference application, load a model, and verify the machine can answer prompts reliably
Read tutorial →