Merge pull request #49681 from GraemeMalcolm/main

Updates to model catalog and other content

Author: JillGrant615

learn-pr/wwl-data-ai/explore-models-azure-ai-studio/1-introduction.yml

@@ -6,7 +6,7 @@ metadata:
   prefetch-feature-rollout: true
   title: Introduction
   description: "Introduction to configuring the deployment of a language model using the Azure AI Foundry portal."
-  ms.date: 03/14/2025
+  ms.date: 03/26/2025
   author: madiepev
   ms.author: madiepev
   ms.topic: unit

learn-pr/wwl-data-ai/explore-models-azure-ai-studio/2-select-model.yml

@@ -1,12 +1,12 @@
 ### YamlMime:ModuleUnit
 uid: learn.wwl.explore-models-copilot-ai-studio.select-model
-title: Explore the language models in the model catalog 
+title: Explore the model catalog 
 metadata:
   adobe-target: true
   prefetch-feature-rollout: true
-  title: Explore the language models in the model catalog
-  description: "Learn how to navigate the model catalog and to select a language model."
-  ms.date: 03/14/2025
+  title: Explore the model catalog
+  description: "Learn how to browse and search the model catalog and to select a language model."
+  ms.date: 03/26/2025
   author: madiepev
   ms.author: madiepev
   ms.topic: unit

learn-pr/wwl-data-ai/explore-models-azure-ai-studio/3-deploy-model.yml

@@ -6,7 +6,7 @@ metadata:
   prefetch-feature-rollout: true
   title: Deploy a model to an endpoint
   description: "Deploy and test a language model in the Azure AI Foundry portal."
-  ms.date: 03/14/2025
+  ms.date: 03/26/2025
   author: madiepev
   ms.author: madiepev
   ms.topic: unit

learn-pr/wwl-data-ai/explore-models-azure-ai-studio/4-improve-model.yml

@@ -1,12 +1,12 @@
 ### YamlMime:ModuleUnit
 uid: learn.wwl.explore-models-copilot-ai-studio.improve-model
-title: Improve the performance of a language model
+title: Optimize model performance
 metadata:
   adobe-target: true
   prefetch-feature-rollout: true
-  title: Improve the performance of a language model
+  title: Optimize model performance
   description: "Explore the various strategies to improve the model's performance"
-  ms.date: 03/14/2025
+  ms.date: 03/26/2025
   author: madiepev
   ms.author: madiepev
   ms.topic: unit

learn-pr/wwl-data-ai/explore-models-azure-ai-studio/5-exercise.yml

@@ -6,7 +6,7 @@ metadata:
   prefetch-feature-rollout: true
   title: Exercise - Explore, deploy, and chat with language models
   description: "Exercise - Explore, deploy, and chat with language models in the Azure AI Foundry portal."
-  ms.date: 03/14/2025
+  ms.date: 03/26/2025
   author: madiepev
   ms.author: madiepev
   ms.topic: unit

learn-pr/wwl-data-ai/explore-models-azure-ai-studio/6-knowledge-check.yml

@@ -6,14 +6,25 @@ metadata:
   prefetch-feature-rollout: true
   title: Knowledge check
   description: "Knowledge check about configuring language models."
-  ms.date: 03/14/2025
+  ms.date: 03/26/2025
   author: madiepev
   ms.author: madiepev
   ms.topic: unit
 durationInMinutes: 3
 content: |
 quiz:
   questions:
+  - content: "Where can you test a deployed model in the Azure AI Foundry portal?"
+    choices:
+    - content: "Chat playground"
+      isCorrect: true
+      explanation: "Correct. The chat playground is designed for interactive testing and experimentation with deployed models, allowing you to explore how the model responds in real-time."
+    - content: "Sandbox"
+      isCorrect: false
+      explanation: "Incorrect. A sandbox is typically a safe environment for testing code changes, but it isn't designed for testing deployed language models."
+    - content: "Development toolbox"
+      isCorrect: false
+      explanation: "Incorrect. The development toolbox can contain various tools for development purposes, but it isn't the place where you would test a deployed model."
   - content: "You want to specify the tone, format, and content for each interaction with your model in the playground. What should you use to customize the model response?"
     choices:
     - content: "Benchmarks"
@@ -25,26 +36,15 @@ quiz:
     - content: "System message"
       isCorrect: true
       explanation: "Correct. System messages are used to customize the model's responses by specifying the tone, format, and content for each interaction."
-  - content: "What pattern can you use to ground prompts?"
+  - content: "What deployment options for an OpenAI model incur no cost for deployment and charge for inferencing based on token-based billing?"
     choices:
-    - content: "Metadata Optimized Prompt (MOP)"
+    - content: "Only Azure OpenAI Service"
       isCorrect: false
-      explanation: "Incorrect. MOP isn't a recognized pattern for grounding prompts."
-    - content: "Retrieval Augmented Generation (RAG)"
-      isCorrect: true
-      explanation: "Correct. RAG involves using a retrieval-based approach to provide relevant information to the model, which can help ground the prompts in specific knowledge or data."
-    - content: "Data Understanding Support Text (DUST)"
+      explanation: "Incorrect. Azure OpenAI Service is not the only option for the required deployment criteria."
+    - content: "Only managed compute"
       isCorrect: false
-      explanation: "Incorrect. DUST isn't a recognized pattern for grounding prompts."
-  - content: "Where can you test a deployed model in the Azure AI Foundry portal?"
-    choices:
-    - content: "Chat playground"
+      explanation: "Incorrect. Managed compute is charged per-minute."
+    - content: "Either Azure OpenAI Service or Azure AI Model Inferencing Service"
       isCorrect: true
-      explanation: "Correct. The chat playground is designed for interactive testing and experimentation with deployed models, allowing you to explore how the model responds in real-time."
-    - content: "Sandbox"
-      isCorrect: false
-      explanation: "Incorrect. A sandbox is typically a safe environment for testing code changes, but it isn't designed for testing deployed language models."
-    - content: "Development toolbox"
-      isCorrect: false
-      explanation: "Incorrect. The development toolbox can contain various tools for development purposes, but it isn't the place where you would test a deployed model."
+      explanation: "Correct. You can deploy OpenAI mdoels to either Azure OpenAI Service or Azure AI Model Inferencing Service."
 

learn-pr/wwl-data-ai/explore-models-azure-ai-studio/7-summary.yml

@@ -6,7 +6,7 @@ metadata:
   prefetch-feature-rollout: true
   title: Summary
   description: "Summary of configuring language models."
-  ms.date: 03/14/2025
+  ms.date: 03/26/2025
   author: madiepev
   ms.author: madiepev
   ms.topic: unit

learn-pr/wwl-data-ai/explore-models-azure-ai-studio/includes/1-introduction.md

@@ -1,8 +1,6 @@
-Foundation models, such as GPT-4, are state-of-the-art natural language processing models designed to understand, generate, and interact with human language. To understand the significance of foundation models, it's essential to explore their origins, which stem from advancements in the field of natural language processing.
+Generative AI applications are built on *language models*. The development process usually starts with an exploration and comparison of available *foundation* models to find the one that best suits the particular needs of your application. After selecting a suitable model, you deploy it to an endpoint where it can be consumed by a client application or AI agent.
 
-## Understand natural language processing
-
-Natural language processing (NLP) is a type of artificial intelligence (AI) that focuses on understanding, interpreting, and generating human language. Some common NLP use cases are:
+Foundation models, such as the GPT family of models, are state-of-the-art language models designed to understand, generate, and interact with natural language. Some common use cases for models are:
 
 - **Speech-to-text and text-to-speech conversion**. For example, generate subtitles for videos.
 - **Machine translation**. For example, translate text from English to Japanese.
@@ -12,16 +10,11 @@ Natural language processing (NLP) is a type of artificial intelligence (AI) that
 - **Question answering**. For example, provide answers to questions like "What is the capital of France?"
 - **Reasoning**. For example, solve a mathematical problem.
 
-> [!Note]
-> In this module, you focus on exploring foundation models used for question answering. The foundation models you explore can be used for chat applications in which you use a language model to generate a response to a user's question.
-
-### Understand the importance of the Transformer architecture
-
-The latest breakthrough in **Natural Language Processing** (**NLP**) is owed to the development of the **Transformer** architecture.
-
-Transformers were introduced in the [*Attention is all you need* paper by Vaswani, et al. from 2017](https://arxiv.org/abs/1706.03762?azure-portal=true). The Transformer architecture provided two innovations to NLP that resulted in the emergence of foundation models:
+In this module, you focus on exploring foundation models used for question answering. The foundation models you explore can be used for chat applications in which you use a language model to generate a response to a user's question.
 
-- Instead of processing words sequentially, Transformers process each word independently and in parallel by using **attention**.
-- Next to the semantic similarity between words, Transformers use **positional encoding** to include the information about the position of a word in a sentence.
+> [!NOTE]
+> The latest breakthrough in generative AI models is owed to the development of the **Transformer** architecture. Transformers were introduced in the [*Attention is all you need* paper by Vaswani, et al. from 2017](https://arxiv.org/abs/1706.03762?azure-portal=true). The Transformer architecture provided two innovations to NLP that resulted in the emergence of foundation models:
+>
+> - Instead of processing words sequentially, Transformers process each word independently and in parallel by using **attention**.
+> - Next to the semantic similarity between words, Transformers use **positional encoding** to include the information about the position of a word in a sentence.
 
-Foundation models designed for NLP use cases are often referred to as **Large Language Models** (**LLMs**) or language models. In this module, you explore the available language models, how to select a model for your use case, and how to use a language model with the Azure AI Foundry portal. You focus on language models that help you to develop generative AI apps that serve as chat applications that do question answering, answering questions to your users.