Local AI Masterclass: LLMs, Diffusion & AI-Agents on Your PC (Udemy.com)

Do you want full control over artificial intelligence?
Learn how to unleash the power of LLMs and image generators directly on your own computer – no cloud, no data risks, maximum performance.

AI automation and intelligent agents are transforming industries overnight.
But what if you don’t want to depend on external providers?
What if you need an AI infrastructure that’s 100% under your control – whether for privacy, unmatched performance, or limitless creative freedom?

This course is your step-by-step guide into the world of local AI.
You won’t just learn how to install individual tools — you’ll build a complete, interconnected ecosystem where language models (LLMs), image and video generators (diffusion models), and intelligent AI agents work seamlessly together.
Everything runs entirely on-premise — right on your own hardware.

We’ll use leading open-source tools like Ollama, LM Studio, Anything LLM, Flowise, n8n, Docker, Supabase, ComfyUI, MCP, and Open WebUI.
Together, we’ll build your personal AI command center for everything — from text automation to stunning image, audio, and video production, and even advanced agent-based workflows.

Forget the limitations of cloud services.
After this course, you’ll be able to build an AI system that surpasses standard solutions like ChatGPT or Stable Diffusion — in functionality, security, and individuality.

WHAT YOU WILL LEARN – SECTION BY SECTION

Section 1: Introduction & Quick Start

• Get a clear overview of the course structure and learning goals.

• Access all the key resources and links you’ll need to succeed.

• Learn how to find the best open-source models for your projects.

• Quick Tip: Install your first local AI app in minutes using Pinokio.

Section 2: Fundamentals – LLMs, Diffusion & Hardware

• Understand the key advantages of local AI in terms of privacy, cost, and performance.

• Demystify hardware requirements — GPU, VRAM, RAM, and unified memory explained simply.

• Discover the best setups for Apple Silicon vs NVIDIA GPUs.

• Grasp how LLMs (GPT, Qwen) and diffusion models (Stable Diffusion, Flux) actually work.

Section 3: Local LLMs in Action – Your Own ChatGPT with Ollama

• Install and configure Ollama from scratch and learn how to use the right models.

• Master prompt engineering, multimodal capabilities (Gemma Vision or Qwen VL), and advanced techniques like Test-Time Compute.

• Integrate external tools via Function Calling & Tool Use to extend your model’s abilities.

• Dive into RAG & embeddings directly in Ollama and optimize with quantization (GGUF, FP8, Q8, Q4 & more).

Section 4: Knowledge Management with Anything LLM & LM Studio (RAG)

• Deeply understand RAG (Retrieval-Augmented Generation), embeddings, and vector databases.

• Build a local RAG chatbot with Anything LLM and Ollama — trained on your own documents.

• Use agent features like web search and scraping, and interact via voice using Whisper.

• Learn how LM Studio serves as a powerful all-in-one alternative to Ollama.

Section 5: Local Diffusion – Image & Video Creation with ComfyUI

• Install and master ComfyUI, the most powerful local image creation tool.

• Understand and use complex JSON workflows for SDXL, Qwen, and Flux.

• Train LoRAs, apply ControlNets for consistent characters, and upscale images professionally.

• Create AI videos, edit images with Qwen Image Edit, and explore workflows via OpenArt.

• Upscale your AI images with ComfyUI using Flux, SDXL, and SUPIR to achieve ultra-sharp, high-resolution results.

Section 6: AI Video Creation on Your PC

• Generate videos directly from text or images using models like Wan.

• Animate characters or replace faces and outfits using Wan Animate.

• Discover advanced local video tools like FramePack.

• Pro Tip: Speed up video generation up to 7× with a custom 4-step LoRA technique.

Section 7: AI Audio – Speech, Voices & Music

• Transcribe audio precisely with Whisper (STT).

• Generate lifelike voices from text (TTS) or clone your own voice with open-source tools.

• Create royalty-free AI music directly on your hardware.

Section 8: Agentic AI – Build Intelligent Automation Agents with n8n

• Install and configure the n8n automation platform locally.

• Set up a local Supabase vector database for your RAG agents via Docker.

• Build a fully functional RAG agent combining Ollama, n8n, and Supabase.

• Connect your agents to SQL databases and external services like Gmail.

Section 9: Advanced Workflows – The Art of System Integration

• Connect everything using MCP as a bridge between your tools.

• Use LM Studio or Open WebUI as your central control hub for AI agents built in n8n.

• Automate image and video creation by linking ComfyUI with n8n.

• Extract data from PDFs, images, and invoices automatically via OCR.

• Explore fine-tuning with Unsloth to customize LLMs for your own data — and understand when the effort truly pays off versus using adapters or RAG.

Section 10: Flowise Crash Course – Build Visual RAG Bots

• Install Flowise and a Postgres database locally with Node and Docker.

• Build a drag-and-drop RAG agent that runs entirely on your own knowledge base.