Teach on Udemy

Turn what you know into an opportunity and reach millions around the world.

Learn More

Your cart is empty.

Keep shopping

Run Local LLMs with Ollama: From No-Code to Python Code

Name: Run Local LLMs with Ollama: From No-Code to Python Code
Rating: 4.6 (98 reviews)

Meta Llama 3, Ollama CLI, OpenWebUI, Multimodal, LangChain, OpenAI Compatibility, LlamaIndex & Function Calling

Created byKshitij Joy (Cloud Alchemy)

Last updated 10/2025

English

English [Auto],

What you'll learn

Basics of Artificial Intelligence
What is Deep Learning ?
What is a Neural Network
LLM Fundamentals
Intro to Meta Llama 3.2/3.3
Llama Vs GPT
How Llama Processes Text : tokens / attention mechanism
Deployment Strategies - HuggingFace , PyTorch, Ollama, Azure with Demos
Deep Dive into Ollama
Installation of Ollama on Linux / Mac / Docker Mode
Demos on Ollama CLI Commands
Building first Custom Model with Ollama & Llama
OpenWebUI - The GUI for Ollama Models
Demos- Using Ollama with VSCode / Colab / Jupyter Notebook
How to use Simple Python Codes in Ollama
MultiModal Capabilities - Analysing Images
LangChain-Ollama-Llama
OpenAI Compatibility
Structure Outputs
Tools & Ollama - Function Calling

Course content

17 sections • 130 lectures • 6h 51m total length

Instruction0:25
Title - LLM Foundations for Absolute Beginners0:08
Introduction1:26
Explore the foundations of large language models for beginners, including deep learning basics, transformer architecture, and quantization, with billions-parameter concepts and deployment groundwork.
What is AI - Artificial Intelligence4:51
AI Vs ML Vs DL3:42
Explore the AI, ML, and DL hierarchy, where AI sits atop, ML learns from data to improve, and DL uses brain-inspired neural networks for driverless cars and generative AI.
What is Deep Learning ?10:02
What is a Neural Network ?9:44
What is LLM ?7:43
Learn what large language models are and how they process input, understand language, and generate human-like text using billions of parameters and transformer-based deep learning.
What are 1B /3B -- Billions of Parameters7:39
What are the Model Benchmarks ?7:00
What are Transformers ?13:03
Understand transformer models as the backbone of generative ai, with self-attention, encoder-decoder blocks, and parallel processing powering text generation, translation, and other natural language processing tasks like BERT and GPT.
What is Embedding ?10:42
What is Quantization ?7:02
What is Context Length of LLM Model ?6:21

Important Information0:25
Title - Intro to Meta Llama0:08
Introduction to Meta Llama1:13
Explore Meta Lama's architecture and capabilities, how the Lama language model powers research and real-world solutions. Preview Meta AI chatbot and artificial analysis quality index arc to evaluate AI performance.
What is Meta Llama ?3:51
Explore what meta llama is, a large language model from Meta AI, and how its natural language processing enables writing, answering questions, and text summarization.
History of Llama6:18
Trace the history of llama models from llama 1 to llama 3.3, highlighting commercial access, code llama, multimodal understanding, and reduced compute at 405 billion parameters.
Llama 3.2 Model8:27
Llama 3.3 Model2:55
Explore llama 3.3, a december 2024 text-only model with 70 billion parameters. It supports eight languages, outputs text and code, and uses a 128k context with 15 trillion tokens.
Differences between LLaMA and other LLMs like GPT5:23
Compare Llama by Meta and GPT by OpenAI, highlighting model sizes from 7 billion to 70 billion, open-source access vs paid API, and tuning options.
How LLaMA processes text: tokens, embeddings, and attention mechanisms4:53
Artificial Analysis Quality Index3:20
Demo: Meta AI Chatbot3:10

Important Information0:25
Title - Deployment Strategies for Meta Llama Models0:08
Introduction - Deployment Strategies0:57
What is Hugging Face ?6:13
Discover hugging face as a leading NLP platform with pre-trained models, datasets, spaces, and the transformers library, offering an inference API for easy predictions and cross-framework deployment.
Demo- Requesting Access for Llama Models3:27
Demo: Running LLaMA models with Hugging Face5:39
What is Pytorch ?4:11
Explore PyTorch, an open source machine learning framework by Facebook's AI research for deep learning and tensor computation, featuring dynamic computational graphs, distributed training across GPUs, and a Python ecosystem.
Demo: Running LLaMA models with PyTorch6:23
Ollama5:10
Explore Ollama’s local deployment of llama models, pulling and running them on your machine with a simple command line interface for private, cost-effective in-house inference.
Demo: Running LLaMA models with Ollama3:09
Cloud Vendors (Azure)4:44
Demo: Running Llama Models with Azure7:18

Title - Introduction to Ollama0:08
Introduction to Ollama0:55
Explore Ollama, a platform that runs large language models locally, tracing its journey to a tool for running AI models and use cases, while highlighting privacy, control, and cloud independence.
What is Ollama ?2:12
Ollama is an open source platform that runs large language models on your local machine or server, enabling a powerful, user friendly AI experience with private data.
History of Ollama3:39
Benefits of Ollama4:00
Use-Case Supported by Ollama3:26
Let's Get Connected0:23

Title - Setting up Ollama0:08
Introduction -Setup Ollama0:51
Explore the Ulama website, review system requirements, and confirm supported operating systems. Follow hands-on demos to install Ulama on macOS, Linux, or via Docker.
Walkthrough of Ollama Website4:48
System Requirements for Ollama3:16
Operating Systems Supported by Ollama1:36
Ollama supports macOS, Linux, and Windows, with macOS 11 Big Sur or later, Linux kernel 4.15 or higher (Ubuntu 18.04+, Fedora 30+, Debian 10+), and Windows 10 or later.
Demo: Installing Ollama on MacOS1:39
Download ollama for macOS, install and move it to applications, launch, and verify with CLI commands like Allama ps and Allama -v showing version 0.5.4.
Demo: Installing Ollama for Linux3:00
Demo: Installing Ollama via Docker7:59

Title - Ollama CLI0:08
Master the Ollama CLI to run and manage models from the command line. The course teaches no-code and Python code workflows from beginner to pro.
Introduction - Ollama CLI0:51
Ollama CLI3:46
Demo: ollama help2:23
Master ollama help to explore command usage quickly. Use optional flags -h and --help to view help for any command, including show and list.
Demo: ollama pull2:29
Demo: ollama run3:11
Demo demonstrates running offline models with llama run, using a downloaded llama three, generating bullet-point outputs, and managing sessions on a laptop or server.
Demo: ollama list1:53
Demo: ollama show3:58
Demo: ollama ps2:18
Learn how to use ollama ps to list and monitor running models, including unique IDs, sizes, and gpu usage, and distinguish between ps and list commands.
Demo: ollama cp2:00
Learn to use the ollama cp command to copy and rename a model, back it up, and run the new model to answer simple questions like two plus two.
Demo: ollama rm1:29

Title-OpenWebUI0:08
Introduction1:32
Explore how Open Web UI hosts and manages llms locally, integrates with llama for model management, visualization, monitoring, and experimentation, and lets you run a personal chatbot in your browser.
What is Openwebui2:53
Demo: Download Docker desktop1:31
Demo: Run docker command to install openwebui1:56
Demo: Open the web browser & use chatbot2:07
Launch and explore the open web ui container locally by accessing port 3000 in a browser, create an admin account, and query llama models like llama 3.2 latest.

Title - Using Various IDE's0:08
Explore using various integrated development environments to accelerate Ollama workflows, blending no-code techniques with Python code for beginner to pro proficiency.
Introduction - Ollama with Various IDE's1:27
Learn to run llama models locally across Jupyter, Colab, and VS Code by setting up virtual environments, notebooks, and Python demos to enhance AI development workflows.
Setup Ollama with Jupyter Notebook4:37
Setup Ollama with Visual Studio Code5:30
Demo: Run a Sample Python Code3:05
Setup Ollama with Google Collab (Ollama running inside Collab)7:16
Learn to set up Ollama inside Google Colab by installing Colab Xterm, opening a terminal, downloading the llama 3.2 model, and running the Ollama service within the Colab environment.
Demo: Run a Sample Python Code2:01
learn how to run llama 3.2 models in Google Colab by installing the llama python library, importing it, and using the generate function with a prompt to create a story.

Title - Simple Python Codes in Ollama0:08
Explore simple Python codes in Ollama with no-code approaches, guiding beginners to pro proficiency using Python codes.
Introduction - Simple Python Codes1:45
Demo: Setup Environment with Github Copilot5:02
Demo: Using generate with ollama (Ollama.generate)2:23
When Ollama is Running Locally (on Your Computer):
Why Use Ngrok?
Google Colab can’t directly access your local machine.
Ngrok creates a secure public URL to expose your locally running Ollama instance to the internet.
Steps:
Run Ollama Locally:
Start Ollama on your local machine (e.g., running at http://localhost:11434).
Install Ngrok and Start a Tunnel:
Install Ngrok on your machine if not already installed.
Run ngrok http 11434.
Ngrok generates a public URL (e.g., https://xyz123.ngrok.io).
Use the Ngrok URL in Colab:
In your Colab notebook, send API requests to the public Ngrok URL (e.g., https://xyz123.ngrok.io).
When Ollama is Running Remotely (on a Server):
Why Use Ngrok?
If the remote server doesn’t have a public IP or its ports are not accessible, Ngrok can expose it to the internet securely.
Steps:
Run Ollama on the Remote Server:
Start Ollama on the remote server (e.g., running at http://localhost:11434).
Install and Run Ngrok on the Remote Server:
SSH into the server.
Run ngrok http 11434.
Ngrok generates a public URL for the server (e.g., https://abc456.ngrok.io).
Use the Ngrok URL in Colab:
In your Colab notebook, send API requests to the Ngrok URL (e.g., https://abc456.ngrok.io).
Demo: Printing only the required artefacts1:07
Demo: Using ollama.chat2:47
Demo: Show how Streaming works with Ollama3:02
Explore how streaming works in Ollama by enabling stream=true and printing output chunk by chunk, with end and flush controlling formatting.
Demo:Ollama with a Custom Client3:03
Learn to use the Ollama client class to chat with a local llama model, instantiate the client, and call the chat method on a specified host and port.
Demo: Create Embedding in Ollama2:34
Learn to create word embeddings with the Ollama client using the embed method on a Llama 3.2 model, including single and multiple words like dog and puppy.

Requirements

No programming experience needed, we cover No-Code and Low-Code Approach

Description

In this course, Ollama Demystified: No-Code & Python Codes for AI Beginners, we have primarily used Meta Llama as the foundational Large Language Model (LLM). However, the techniques and methodologies covered throughout the course are not limited to Meta Llama alone. The same principles can be seamlessly applied to other open-source models such as DeepSeek, Qwen, and many more.

Our goal is to equip you with a solid understanding of how to work with LLMs using both no-code tools and Python programming, enabling you to adapt these skills to various models based on your specific needs and preferences.

Welcome Learners, Unlock the power of cutting-edge AI with Meta LLaMA 3 in this comprehensive beginner-to-pro course! Whether you're new to AI or looking to deepen your expertise, this course offers a step-by-step guide to mastering Meta’s advanced LLaMA 3 language model using Ollama, an intuitive platform that simplifies working with local LLMs.

You’ll start with the basics, understanding what LLaMA 3 is and how it differs from other AI models. Gradually, you'll dive into hands-on projects that guide you through setup, fine-tuning, and leveraging its capabilities for real-world applications. By the end of the course, you’ll confidently use LLaMA 3 with Ollama to build projects, solve problems, and stay at the forefront of AI innovation.

Who Is This Course For?
This course is designed for:

Beginners eager to explore AI with no prior experience.
Tech enthusiasts who want to understand and use advanced AI models.
Developers aiming to integrate AI into personal or professional projects.

What You will Learn ?

1. Introduction to AI , Neural Networks & LLM

1.1 Introduction
1.2 What is AI - Artificial Intelligence
1.3 AI Vs ML Vs DL
1.4 What is a Neural Network?
1.5 What are 1B/3B - Billions of Parameters
1.6 What are the Model Benchmarks?
1.7 What are Transformers?
1.8 What is Embedding?
1.9 What is Quantization?
1.10 What is Context Length of LLM Model?

2. Introduction to Meta LLaMA

2.1 Title - Intro to Meta LLaMA
2.2 Introduction to Meta LLaMA
2.3 What is Meta LLaMA?
2.4 History of LLaMA
2.5 LLaMA 3.2 Model
2.6 LLaMA 3.3 Model
2.7 Differences between LLaMA and other LLMs like GPT
2.8 How LLaMA processes text: tokens, embeddings, and attention mechanisms
2.9 Artificial Analysis Quality Index
2.10 Demo: Meta AI Chatbot

3. Deployment Strategies for Meta LLaMA Models

3.1 Title - Deployment Strategies for Meta LLaMA Models
3.2 Introduction - Deployment Strategies
3.3 What is Hugging Face?
3.4 Demo: Requesting Access for LLaMA Models
3.5 Demo: Running LLaMA Models with Hugging Face
3.6 What is PyTorch?
3.7 Demo: Running LLaMA Models with PyTorch
3.8 Ollama
3.9 Demo: Running LLaMA Models with Ollama
3.10 Cloud Vendors (Azure)
3.11 Demo: Running LLaMA Models with Azure

4. Introduction to Ollama

4.1 Title - Introduction to Ollama
4.2 Introduction to Ollama
4.3 What is Ollama?
4.4 History of Ollama
4.5 Benefits of Ollama
4.6 Use-Cases Supported by Ollama

5. Setting up Ollama

5.1 Title - Setting up Ollama
5.2 Introduction - Setup Ollama
5.3 Walkthrough of Ollama Website
5.4 System Requirements for Ollama
5.5 Operating Systems Supported by Ollama
5.6 Demo: Installing Ollama on MacOS
5.7 Demo: Installing Ollama for Linux
5.8 Demo: Installing Ollama via Docker

6. Ollama CLI

6.1 Title - Ollama CLI
6.2 Introduction - Ollama CLI
6.3 Ollama CLI Overview
6.4 Demo: ollama help
6.5 Demo: ollama pull
6.6 Demo: ollama run
6.7 Demo: ollama list
6.8 Demo: ollama show
6.9 Demo: ollama ps
6.10 Demo: ollama cp
6.11 Demo: ollama rm

7. Building Your Custom Model with Ollama

7.1 Title - Building Your Custom Model with Ollama
7.2 Introduction - Your Own Custom Model
7.3 What is a Model File?
7.4 Demo: Understanding the Contents of a Model File
7.5 Demo: Create Your Custom Model
7.6 Demo: User Interaction
7.7 Demo: Create Custom Model using GGUF File

8. OpenWebUI

8.1 Title - OpenWebUI
8.2 Introduction
8.3 What is OpenWebUI?
8.4 Demo: Download Docker Desktop
8.5 Demo: Run Docker Command to Install OpenWebUI
8.6 Demo: Open the Web Browser & Use Chatbot

9. Using Various IDEs

9.1 Title - Using Various IDEs
9.2 Introduction - Ollama with Various IDEs
9.3 Setup Ollama with Jupyter Notebook
9.4 Setup Ollama with Visual Studio Code
9.5 Demo: Run a Sample Python Code
9.6 Setup Ollama with Google Colab
9.7 Demo: Run a Sample Python Code in Colab

10. Simple Python Codes in Ollama

10.1 Title - Simple Python Codes in Ollama
10.2 Introduction - Simple Python Codes
10.3 Demo: Setup Environment with GitHub Copilot
10.4 Demo: Using ollama.generate
10.5 Demo: Printing Required Artifacts
10.6 Demo: Using ChatOllama
10.7 Demo: Show Streaming with Ollama
10.8 Demo: Ollama with a Custom Client
10.9 Demo: Create Embedding in Ollama

11. Ollama & Multimodality

11.1 Title - Ollama & Multimodality
11.2 Introduction - Multimodal Models
11.3 What is Meta LLaMA 3.2 Vision Model?
11.4 Demo: Analyze an Image Using Ollama CLI

12. LangChain with Ollama & LLaMA

12.1 Title - LangChain with Ollama & LLaMA
12.2 Introduction - Ollama & LangChain
12.3 What is LangChain?
12.4 Ollama with LangChain - ChatOllama
12.5 Demo: Setup Environment for LangChain Work
12.6 Demo: A Simple Python Code with Ollama & LangChain
12.7 Demo: Show the Chaining Concept in LangChain
12.8 Demo: Increase the Level of Chaining, Convert Output to String

13. Ollama & OpenAI Compatibility

13.1 Title - Ollama & OpenAI Compatibility
13.2 Introduction - Ollama & OpenAI Compatibility
13.3 What is OpenAI?
13.4 What is the Ollama & OpenAI Compatibility?
13.5 Demo: How to Get the Same Code Working for Ollama

14. Getting Structured Outputs

14.1 Title - Getting Structured Outputs
14.2 Introduction to Structured Outputs
14.3 What are Structured Outputs with Ollama?
14.4 Demo: Python Code for Structured Output
14.5 Demo: Python Code to Get Objects in JSON Format from an Image

15. Tools in LLaMA & Ollama

15.1 Title - Tools in LLaMA & Ollama
15.2 Introduction to Tools
15.3 What are Tools in Ollama?
15.4 Demo: Understand the Workflow
15.5 Demo: Create an API Key in OpenWeatherMap
15.6 Demo: Using Tools and Function Calling

Who this course is for:

Beginners to Generative AI
Beginners to Meta Llama
Beginners to Ollama
Enthusiasts who wish to learn Ollama with Llama
Professionals wishing to understand and see demos on Ollama Python Library

Run Local LLMs with Ollama: From No-Code to Python Code

What you'll learn

Explore related topics

Course content

Large Language Models (LLM) Foundations (For Absolute Beginners)14 lectures • 1hr 30min

Intro to Meta Llama11 lectures • 40min

Deployment Strategies for Meta's LLaMA Models12 lectures • 48min

Introduction to Ollama7 lectures • 15min

Setting Up Ollama8 lectures • 23min

Understanding Ollama Commands - CLI11 lectures • 24min

Building Your First Custom Model with Ollama & Llama6 lectures • 14min

OpenWebUI - The UI to Llama Models6 lectures • 10min

Use Llama Models with Ollama on Various IDE's (Jupyter / Colab / VSCode)7 lectures • 24min

Simple Python Codes in Ollama9 lectures • 22min

Requirements

Description

Who this course is for: