Azure AI for Developers: LLMs and SLMs
1h 19mIntermediate2025-04-11
Authors
Sam Nasr
Course details
As an Azure developer, it’s important to stay up to date with the latest technological advancements in your field. In this course, instructor Sam Nasr covers the fundamental concepts of small language models (SLMs) and large language models (LLMs), equipping you with the core skills required to leverage the power of AI. Explore the key differences between each model, how they work, and when it's appropriate to use each and why. Along the way, Sam also covers best practices for improving efficiency and shows you how to avoid common mistakes.
Learning objectives
Describe a small language model (SLM) and a large language model (LLM).
Identify the key differences between a SLM and a LLM.
Determine circumstances under which it is most appropriate to use each model type.
Learning objectives
Describe a small language model (SLM) and a large language model (LLM).
Identify the key differences between a SLM and a LLM.
Determine circumstances under which it is most appropriate to use each model type.
Skills covered
Azure AI ServicesCloud AdministrationGenerative AICloud PlatformsArtificial Intelligence (AI)Cloud ComputingMicrosoftDeep Dive (X:Y)
Concepts
0. Introduction
- 01 - Choosing the right AI model - LLMs and SLMs explained
- 02 - What you should know
1. Small Language Models
- 03 - What is a small language model (SLM)
- 04 - How do SLMs work
- 05 - Popular SLM architectures
- 06 - Capabilities and limitations
2. Large Language Models
- 07 - What is a large language model (LLM)
- 08 - How do LLMs work
- 09 - Popular LLM architectures
- 10 - Capabilities and limitations
3. Choosing a Model
- 11 - Difference between SLM and LLM
- 12 - When to use SLM or LLM
- 13 - Deployment types
4. Sample Projects
- 14 - Deploying and accessing SLM locally
- 15 - Deploying and accessing LLM in Azure AI Foundry
- 16 - Challenge - Build an SLM solution
- 17 - Solution - Build an SLM solution
5. Implementing Best Practices
- 18 - Common pitfalls
- 19 - Best practices
Conclusion
- 20 - Next steps