14.1 Customizing Prompt Settings and Models

Microsoft Certified: Intelligent Applications Builder Associate AB-410

Track your progress as you read

Finished sections get a check mark.

1 Introduction

1.1 About the AB-410 Exam 1.2 How to Use This Study Guide

2 Chapter 2: Designing Power Platform Solutions with AI-Enabled Tools

2.1 Analyzing Requirements and Identifying Components 2.2 Evaluating Built-in Agents for Business Solutions 2.3 Recommending Extensibility Options

3 Chapter 3: Environments and ALM Strategy

3.1 Recommending Environment Types 3.2 Solution and Application Lifecycle Management Strategy

4 Chapter 4: Dataverse Tables and Columns

4.1 Creating and Editing Tables in the Data Workspace 4.2 Configuring Table Properties 4.3 Creating and Modifying Columns

5 Chapter 5: Relationships, AI Data Features, and Data Security

5.1 Configuring Table Relationships 5.2 Configuring Prompt Columns and Row Summaries 5.3 Configuring Public Views and Main Forms 5.4 Configuring Dataverse Security

6 Chapter 6: Model-Driven App Forms and Views

6.1 Creating and Configuring Forms for Model-Driven Apps 6.2 Creating and Configuring Views for Model-Driven Apps 6.3 Creating Generative Pages with Natural Language

7 Chapter 7: Composing, Securing, and Visualizing Model-Driven Apps

7.1 Composing a Model-Driven App 7.2 Configuring Access to Model-Driven Apps, Forms, and Views 7.3 Model-Driven App Charts and Dashboards

8 Chapter 8: Creating and Designing Canvas Apps

8.1 Creating a Canvas App Using Data 8.2 Designing for Accessibility, Performance, Responsiveness, and Usability 8.3 Automating Business Processes from Canvas Apps

9 Chapter 9: Canvas App Components, Variables, and Error Handling

9.1 Reusable Components, Named Formulas, and User-Defined Functions 9.2 Managing Variables and Collections 9.3 Implementing Error Handling in Canvas Apps

10 Chapter 10: Testing Canvas Apps and Creating Agents from Apps

10.1 Testing Canvas Apps with Monitor 10.2 Creating a Copilot Studio Agent from a Canvas App

11 Chapter 11: Cloud Flow Triggers, Connectors, and Approvals

11.1 Recommending Cloud Flow Triggers 11.2 Evaluating and Recommending Connectors 11.3 Managing Approvals with Cloud Flows

12 Chapter 12: Cloud Flow Actions, Control, and Troubleshooting

12.1 Configuring Cloud Flow Actions 12.2 Implementing Flow Control with Conditions and Loops 12.3 Testing and Troubleshooting Cloud Flows

13 Chapter 13: Building Prompts in AI Hub

13.1 Building Prompts from Templates or from Blank 13.2 Adding Inputs to a Prompt 13.3 Adding Knowledge to a Prompt

14 Chapter 14: Prompt Settings, Models, and Consuming AI

14.1 Customizing Prompt Settings and Models 14.2 Consuming Prompts in Apps and Cloud Flows 14.3 Consuming AI Models in Apps and Cloud Flows

15 Chapter 15: Business Rules and Business Process Flows

15.1 Configuring Business Rules 15.2 Configuring Business Process Flows

16 Chapter 16: Calculated Columns and Business Logic Strategy

16.1 Calculated, Rollup, and Formula Columns 16.2 Evaluating Use Cases for Business Logic

14.1 Customizing Prompt Settings and Models

Key Takeaways Explain This Topic AI

The prompt builder Model dropdown defaults to GPT-4.1 mini (Basic licensing rate); GPT-4.1, GPT-5 chat, and GPT-5.3 chat are Standard rate, and GPT-5 / GPT-5.2 reasoning models are Premium rate

Temperature is a 0 to 1 slider (default 0) controlling determinism versus creativity, and it is disabled for GPT-5 reasoning models

Prompt output format can be Text or JSON; a custom JSON schema is defined by pasting an example, and JSON output became generally available in November 2024

Content moderation (Low, Moderate default, High) is only adjustable for managed models — the slider is unavailable for Anthropic or Azure AI Foundry models

Prompts used in Power Apps and Power Automate consume AI Builder credits first, then Copilot Credits; prompts used in Copilot Studio consume Copilot Credits only

Last updated: July 2026

Every custom prompt you build in AI hub > Prompts (from make.powerapps.com or make.powerautomate.com) runs on a generative AI model you can choose and tune. The AB-410 exam expects you to know which levers exist in the prompt builder , where they live, and — most importantly — which tradeoffs each one introduces. Two places matter: the Model dropdown at the top of the prompt builder, and the settings panel reached by selecting the three dots ( … ) > Settings .

Selecting the underlying model

Select Model at the top of the prompt builder to pick the model that generates answers. As of mid-2026 the lineup is organized into three categories, each with a different licensing rate and behavior profile:

Model Category Licensing rate Context Notes
GPT-4.1 mini (default) Mini Basic 128K tokens Default model; best cost/speed balance
GPT-4.1 General Standard 128K tokens Complex, multimodal tasks
GPT-5 chat General Standard 128K tokens Newer general model
GPT-5.3 chat General Standard 128K tokens Managed model
GPT-5 reasoning Deep Premium 400K tokens Reasons before responding; slower
GPT-5.2 reasoning Deep Premium 400K tokens Latest deep reasoning model

Claude Sonnet 4.6 / Opus 4.6 General / Deep Standard / Premium 200K tokens External Anthropic models, hosted outside Microsoft

Exam trap: the older GPT-4o mini and GPT-4o names still appear in legacy AI Builder documentation and remain in use in US government regions ; on the commercial cloud the current defaults are the GPT-4.1 family. If a question describes a default, cost-effective model, think "the mini model" rather than memorizing a version number.

Models also carry release stages : Experimental (not for production), Preview (not yet recommended for production), no tag ( generally available ), and Default (the recommended GA model, also used as fallback if your selected model is turned off). Experimental and preview models can show variability in performance, latency, and availability — never the right answer for a production scenario on the exam. Anthropic Claude models count as experimental even though they show no tag, and experimental external models like Grok 4.1 Fast are explicitly not recommended for production use.

Category guidance
Mini — fast and cost-effective; summarization, information tasks, routine document processing
General — superior quality for complex or multimodal work; advanced content creation

Deep — trained for reasoning; data analysis and multi-step problem solving, but slower because the model reasons before responding and bills at the Premium rate

Temperature

The temperature slider (0 to 1, default 0 ) controls randomness. At 0 the model produces predictable, consistent, conservative output — what you want for extraction, classification, and compliance-sensitive text. At 1 you get diverse, creative responses — appropriate for brainstorming or marketing copy. Temperature influences output but never guarantees it; responses remain probabilistic even at the same setting. Exam trap: the temperature slider is disabled for GPT-5 reasoning models — if a scenario demands both deep reasoning and a tunable temperature, that combination does not exist.

Output format: text versus JSON

Under prompt settings you can set the Output to Text or JSON . Choosing JSON lets you define a custom format by pasting an example JSON structure, and the model's reply is constrained to that shape (generally available since November 2024). Use text output for prose shown to a human; use JSON output whenever downstream Power Fx logic or a cloud flow must read individual fields — it eliminates fragile string parsing and pairs with the Parse JSON action in Power Automate.

Remaining settings
Record retrieval — how many records are pulled from attached knowledge sources when grounding the prompt
Include links in the response — adds link citations for retrieved records
Enable code interpreter — lets the model generate and execute code, useful for calculation-heavy answers

Content moderation level — Low , Moderate (default), or High ; stricter levels filter more potentially harmful content but answer fewer prompts. Exam trap: this slider is available only for managed models — it is unavailable when you select Anthropic or Azure AI Foundry models.

Tradeoffs and when to switch

Switching models is a cost/latency/reliability decision. A support-summarization prompt processing thousands of emails a day belongs on the mini model at the Basic rate; moving it to a Deep model multiplies credit consumption and slows every call. Conversely, a month-end variance-analysis prompt that must reason across large tables justifies the Premium rate and extra latency of a reasoning model. Keep temperature at 0 for anything deterministic, and pin production prompts to GA models only.

Licensing follows the same pattern: prompts run in Power Apps or Power Automate consume AI Builder credits first , then Copilot Credits ; prompts run in Copilot Studio consume Copilot Credits only. Rate tier (Basic/Standard/Premium) scales consumption with the model you chose — so model selection is also the primary cost-control lever.