Microsoft AI-103 Image And Video Generation And Editing Practice Test

 

This practice topic covers Image and Video Generation and Editing for Microsoft AI-103. Questions are original and aligned to the current Microsoft skills outline. For broader exam preparation, review the Microsoft AI-103 Exam Dumps page.

Question 1

A runtime design must distinguish text-only image generation from reference-conditioned creation. Which implementation uses the correct capability for the requirement?

1. Use a preview capability for a workload whose production policy prohibits previews.
2. Use reference-conditioned generation when the output must preserve characteristics of a supplied image.
3. Ignore the selected API contract when interpreting an edit mask.
4. Use a text-only model and rely on post-processing to invent missing visual content.
5. Regenerate the full image for every small edit instead of using the supported editing input.

Correct Answer: B

Correct Answer

Answer B is correct because this choice directly implements the required decision for text-only image generation versus reference-conditioned creation. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement text-only image generation versus reference-conditioned creation; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement text-only image generation versus reference-conditioned creation; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would use a text-only model and rely on post-processing to invent missing visual content. That can address a neighboring concern, but it does not implement text-only image generation versus reference-conditioned creation; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement text-only image generation versus reference-conditioned creation; the decisive requirement in the stem therefore remains unresolved.

 

Question 2

In a production AI solution, the following condition occurs: subject and background must be controlled. Which action should the engineer take to handle prompt composition correctly?

1. Write the prompt so subject, background, composition, and required constraints are described separately and unambiguously.
2. Regenerate the full image for every small edit instead of using the supported editing input.
3. Use a preview capability for a workload whose production policy prohibits previews.
4. Use a text-only model and rely on post-processing to invent missing visual content.
5. Generate at an arbitrary aspect ratio and crop essential content afterward.

Correct Answer: A

Correct Answer

Answer A is correct because this choice directly implements the required decision for prompt composition when subject and background must be controlled. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer B is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement prompt composition when subject and background must be controlled; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement prompt composition when subject and background must be controlled; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would use a text-only model and rely on post-processing to invent missing visual content. That can address a neighboring concern, but it does not implement prompt composition when subject and background must be controlled; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would generate at an arbitrary aspect ratio and crop essential content afterward. That can address a neighboring concern, but it does not implement prompt composition when subject and background must be controlled; the decisive requirement in the stem therefore remains unresolved.

 

Question 3

In a production AI solution, the following condition occurs: visual identity must be preserved. Which action should the engineer take to handle reference image handling correctly?

1. Assume prompt wording guarantees exact pixel placement without validating the output.
2. Generate at an arbitrary aspect ratio and crop essential content afterward.
3. Use a text-only model and rely on post-processing to invent missing visual content.
4. Regenerate the full image for every small edit instead of using the supported editing input.
5. Provide the reference image through the supported conditioning mechanism when visual identity must be preserved.

Correct Answer: E

Correct Answer

Answer E is correct because this choice directly implements the required decision for reference image handling when visual identity must be preserved. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement reference image handling when visual identity must be preserved; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would generate at an arbitrary aspect ratio and crop essential content afterward. That can address a neighboring concern, but it does not implement reference image handling when visual identity must be preserved; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would use a text-only model and rely on post-processing to invent missing visual content. That can address a neighboring concern, but it does not implement reference image handling when visual identity must be preserved; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement reference image handling when visual identity must be preserved; the decisive requirement in the stem therefore remains unresolved.

 

Question 4

In a production AI solution, the following condition occurs: the API returns encoded content. Which action should the engineer take to handle image result decoding correctly?

1. Use a text-only model and rely on post-processing to invent missing visual content.
2. Regenerate the full image for every small edit instead of using the supported editing input.
3. Ignore the selected API contract when interpreting an edit mask.
4. Decode the returned image payload according to the response contract before storing or displaying it.
5. Choose a video model for a single still-image requirement only because it uses more compute.

Correct Answer: D

Correct Answer

Answer D is correct because this choice directly implements the required decision for image result decoding when the API returns encoded content. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would use a text-only model and rely on post-processing to invent missing visual content. That can address a neighboring concern, but it does not implement image result decoding when the API returns encoded content; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement image result decoding when the API returns encoded content; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement image result decoding when the API returns encoded content; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would choose a video model for a single still-image requirement only because it uses more compute. That can address a neighboring concern, but it does not implement image result decoding when the API returns encoded content; the decisive requirement in the stem therefore remains unresolved.

 

Question 5

In a production AI solution, the required outcome is a text-only brief. Which action should the engineer take to handle video generation input choice correctly?

1. Use a text-only model and rely on post-processing to invent missing visual content.
2. Choose a video model for a single still-image requirement only because it uses more compute.
3. Use the supported text-to-video input path when the brief contains no reference image or source video.
4. Ignore the selected API contract when interpreting an edit mask.
5. Generate at an arbitrary aspect ratio and crop essential content afterward.

Correct Answer: C

Correct Answer

Answer C is correct because this choice directly implements the required decision for video generation input choice for a text-only brief. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would use a text-only model and rely on post-processing to invent missing visual content. That can address a neighboring concern, but it does not implement video generation input choice for a text-only brief; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would choose a video model for a single still-image requirement only because it uses more compute. That can address a neighboring concern, but it does not implement video generation input choice for a text-only brief; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement video generation input choice for a text-only brief; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would generate at an arbitrary aspect ratio and crop essential content afterward. That can address a neighboring concern, but it does not implement video generation input choice for a text-only brief; the decisive requirement in the stem therefore remains unresolved.

 

Question 6

In a production AI solution, the following condition occurs: the opening appearance is specified. Which action should the engineer take to handle reference-frame use correctly?

1. Assume prompt wording guarantees exact pixel placement without validating the output.
2. Use the supplied reference frame when the opening appearance must be preserved.
3. Discard the reference image even though subject consistency is required.
4. Regenerate the full image for every small edit instead of using the supported editing input.
5. Generate at an arbitrary aspect ratio and crop essential content afterward.

Correct Answer: B

Correct Answer

Answer B is correct because this choice directly implements the required decision for reference-frame use when the opening appearance is specified. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement reference-frame use when the opening appearance is specified; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement reference-frame use when the opening appearance is specified; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement reference-frame use when the opening appearance is specified; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would generate at an arbitrary aspect ratio and crop essential content afterward. That can address a neighboring concern, but it does not implement reference-frame use when the opening appearance is specified; the decisive requirement in the stem therefore remains unresolved.

 

Question 7

In a production AI solution, the required outcome is a long-running video request. Which action should the engineer take to handle asynchronous job handling correctly?

1. Submit the long-running video request asynchronously and poll or receive completion status before retrieving the result.
2. Regenerate the full image for every small edit instead of using the supported editing input.
3. Ignore the selected API contract when interpreting an edit mask.
4. Assume prompt wording guarantees exact pixel placement without validating the output.
5. Use a preview capability for a workload whose production policy prohibits previews.

Correct Answer: A

Correct Answer

Answer A is correct because this choice directly implements the required decision for asynchronous job handling for a long-running video request. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer B is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement asynchronous job handling for a long-running video request; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement asynchronous job handling for a long-running video request; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement asynchronous job handling for a long-running video request; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement asynchronous job handling for a long-running video request; the decisive requirement in the stem therefore remains unresolved.

 

Question 8

A multimodal workload requires failed generation handling without returning an incomplete artifact. Which implementation best preserves the required visual evidence and policy boundary?

1. Use a text-only model and rely on post-processing to invent missing visual content.
2. Discard the reference image even though subject consistency is required.
3. Ignore the selected API contract when interpreting an edit mask.
4. Choose a video model for a single still-image requirement only because it uses more compute.
5. Return a clear generation failure instead of an incomplete artifact.

Correct Answer: E

Correct Answer

Answer E is correct because this choice directly implements the required decision for failed generation handling without returning an incomplete artifact. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would use a text-only model and rely on post-processing to invent missing visual content. That can address a neighboring concern, but it does not implement failed generation handling without returning an incomplete artifact; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement failed generation handling without returning an incomplete artifact; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement failed generation handling without returning an incomplete artifact; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would choose a video model for a single still-image requirement only because it uses more compute. That can address a neighboring concern, but it does not implement failed generation handling without returning an incomplete artifact; the decisive requirement in the stem therefore remains unresolved.

 

Question 9

A runtime design must distinguish localized inpainting from regenerating the whole image. Which implementation uses the correct capability for the requirement?

1. Generate at an arbitrary aspect ratio and crop essential content afterward.
2. Assume prompt wording guarantees exact pixel placement without validating the output.
3. Use a preview capability for a workload whose production policy prohibits previews.
4. Use localized inpainting when only the selected image region should change.
5. Discard the reference image even though subject consistency is required.

Correct Answer: D

Correct Answer

Answer D is correct because this choice directly implements the required decision for localized inpainting versus regenerating the whole image. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would generate at an arbitrary aspect ratio and crop essential content afterward. That can address a neighboring concern, but it does not implement localized inpainting versus regenerating the whole image; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement localized inpainting versus regenerating the whole image; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement localized inpainting versus regenerating the whole image; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement localized inpainting versus regenerating the whole image; the decisive requirement in the stem therefore remains unresolved.

 

Question 10

In a production AI solution, the workload includes the editable source image. Which action should the engineer take to handle mask dimensions aligned correctly?

1. Discard the reference image even though subject consistency is required.
2. Assume prompt wording guarantees exact pixel placement without validating the output.
3. Match the edit mask dimensions to the editable source image.
4. Regenerate the full image for every small edit instead of using the supported editing input.
5. Ignore the selected API contract when interpreting an edit mask.

Correct Answer: C

Correct Answer

Answer C is correct because this choice directly implements the required decision for mask dimensions aligned with the editable source image. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement mask dimensions aligned with the editable source image; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement mask dimensions aligned with the editable source image; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement mask dimensions aligned with the editable source image; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement mask dimensions aligned with the editable source image; the decisive requirement in the stem therefore remains unresolved.

 

Question 11

A multimodal workload requires mask polarity checked against the selected API contract. Which implementation best preserves the required visual evidence and policy boundary?

1. Assume prompt wording guarantees exact pixel placement without validating the output.
2. Interpret mask polarity according to the selected API contract.
3. Choose a video model for a single still-image requirement only because it uses more compute.
4. Use a preview capability for a workload whose production policy prohibits previews.
5. Discard the reference image even though subject consistency is required.

Correct Answer: B

Correct Answer

Answer B is correct because this choice directly implements the required decision for mask polarity checked against the selected API contract. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement mask polarity checked against the selected API contract; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would choose a video model for a single still-image requirement only because it uses more compute. That can address a neighboring concern, but it does not implement mask polarity checked against the selected API contract; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement mask polarity checked against the selected API contract; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement mask polarity checked against the selected API contract; the decisive requirement in the stem therefore remains unresolved.

 

Question 12

A multimodal workload requires prompt-driven edit that preserves unselected regions. Which implementation best preserves the required visual evidence and policy boundary?

1. Apply the edit only to the selected region and preserve unselected regions.
2. Ignore the selected API contract when interpreting an edit mask.
3. Choose a video model for a single still-image requirement only because it uses more compute.
4. Use a text-only model and rely on post-processing to invent missing visual content.
5. Generate at an arbitrary aspect ratio and crop essential content afterward.

Correct Answer: A

Correct Answer

Answer A is correct because this choice directly implements the required decision for prompt-driven edit that preserves unselected regions. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer B is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement prompt-driven edit that preserves unselected regions; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would choose a video model for a single still-image requirement only because it uses more compute. That can address a neighboring concern, but it does not implement prompt-driven edit that preserves unselected regions; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would use a text-only model and rely on post-processing to invent missing visual content. That can address a neighboring concern, but it does not implement prompt-driven edit that preserves unselected regions; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would generate at an arbitrary aspect ratio and crop essential content afterward. That can address a neighboring concern, but it does not implement prompt-driven edit that preserves unselected regions; the decisive requirement in the stem therefore remains unresolved.

 

Question 13

A workload has a specific constraint: under a stated requirement. The implementation must choose correctly between reference fidelity and creative modification. Which option best satisfies the constraint?

1. Ignore the selected API contract when interpreting an edit mask.
2. Assume prompt wording guarantees exact pixel placement without validating the output.
3. Discard the reference image even though subject consistency is required.
4. Use a text-only model and rely on post-processing to invent missing visual content.
5. Prioritize reference fidelity when the requirement says the supplied subject/style must be preserved.

Correct Answer: E

Correct Answer

Answer E is correct because this choice directly implements the required decision for reference fidelity versus creative modification under a stated requirement. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement reference fidelity versus creative modification under a stated requirement; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement reference fidelity versus creative modification under a stated requirement; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement reference fidelity versus creative modification under a stated requirement; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would use a text-only model and rely on post-processing to invent missing visual content. That can address a neighboring concern, but it does not implement reference fidelity versus creative modification under a stated requirement; the decisive requirement in the stem therefore remains unresolved.

 

Question 14

A runtime design must distinguish targeted remix of a generated clip from full regeneration. Which implementation uses the correct capability for the requirement?

1. Use a preview capability for a workload whose production policy prohibits previews.
2. Regenerate the full image for every small edit instead of using the supported editing input.
3. Ignore the selected API contract when interpreting an edit mask.
4. Use a targeted remix/edit when only a bounded video segment needs modification.
5. Assume prompt wording guarantees exact pixel placement without validating the output.

Correct Answer: D

Correct Answer

Answer D is correct because this choice directly implements the required decision for targeted remix of a generated clip versus full regeneration. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement targeted remix of a generated clip versus full regeneration; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement targeted remix of a generated clip versus full regeneration; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement targeted remix of a generated clip versus full regeneration; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement targeted remix of a generated clip versus full regeneration; the decisive requirement in the stem therefore remains unresolved.

 

Question 15

In a production AI solution, the required outcome is an editing operation. Which action should the engineer take to handle source video identifier required correctly?

1. Choose a video model for a single still-image requirement only because it uses more compute.
2. Use a text-only model and rely on post-processing to invent missing visual content.
3. Pass the required source video identifier to the editing operation for an editing operation.
4. Ignore the selected API contract when interpreting an edit mask.
5. Regenerate the full image for every small edit instead of using the supported editing input.

Correct Answer: C

Correct Answer

Answer C is correct because this choice directly implements the required decision for source video identifier required for an editing operation. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would choose a video model for a single still-image requirement only because it uses more compute. That can address a neighboring concern, but it does not implement source video identifier required for an editing operation; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would use a text-only model and rely on post-processing to invent missing visual content. That can address a neighboring concern, but it does not implement source video identifier required for an editing operation; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement source video identifier required for an editing operation; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement source video identifier required for an editing operation; the decisive requirement in the stem therefore remains unresolved.

 

Question 16

In a production AI solution, the condition appears after a local video edit. Which action should the engineer take to handle temporal continuity verification correctly?

1. Ignore the selected API contract when interpreting an edit mask.
2. Verify temporal continuity across the edited video segment boundaries after a local video edit.
3. Choose a video model for a single still-image requirement only because it uses more compute.
4. Use a preview capability for a workload whose production policy prohibits previews.
5. Discard the reference image even though subject consistency is required.

Correct Answer: B

Correct Answer

Answer B is correct because this choice directly implements the required decision for temporal continuity verification after a local video edit. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement temporal continuity verification after a local video edit; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would choose a video model for a single still-image requirement only because it uses more compute. That can address a neighboring concern, but it does not implement temporal continuity verification after a local video edit; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement temporal continuity verification after a local video edit; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement temporal continuity verification after a local video edit; the decisive requirement in the stem therefore remains unresolved.

 

Question 17

In a production AI solution, the workload operates under a fixed display requirement. Which action should the engineer take to handle supported output dimensions correctly?

1. Choose output dimensions supported by the model that meet the fixed display requirement under a fixed display requirement.
2. Regenerate the full image for every small edit instead of using the supported editing input.
3. Discard the reference image even though subject consistency is required.
4. Choose a video model for a single still-image requirement only because it uses more compute.
5. Generate at an arbitrary aspect ratio and crop essential content afterward.

Correct Answer: A

Correct Answer

Answer A is correct because this choice directly implements the required decision for supported output dimensions under a fixed display requirement. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer B is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement supported output dimensions under a fixed display requirement; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement supported output dimensions under a fixed display requirement; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would choose a video model for a single still-image requirement only because it uses more compute. That can address a neighboring concern, but it does not implement supported output dimensions under a fixed display requirement; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would generate at an arbitrary aspect ratio and crop essential content afterward. That can address a neighboring concern, but it does not implement supported output dimensions under a fixed display requirement; the decisive requirement in the stem therefore remains unresolved.

 

Question 18

A multimodal workload requires quality-versus-latency setting on a verified model. Which implementation best preserves the required visual evidence and policy boundary?

1. Use a preview capability for a workload whose production policy prohibits previews.
2. Regenerate the full image for every small edit instead of using the supported editing input.
3. Ignore the selected API contract when interpreting an edit mask.
4. Discard the reference image even though subject consistency is required.
5. Select the documented quality/latency setting that meets the workload target.

Correct Answer: E

Correct Answer

Answer E is correct because this choice directly implements the required decision for quality-versus-latency setting on a verified model. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement quality-versus-latency setting on a verified model; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement quality-versus-latency setting on a verified model; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would ignore the selected API contract when interpreting an edit mask. That can address a neighboring concern, but it does not implement quality-versus-latency setting on a verified model; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement quality-versus-latency setting on a verified model; the decisive requirement in the stem therefore remains unresolved.

 

Question 19

In a production AI solution, the following condition occurs: transparency is required. Which action should the engineer take to handle output format choice correctly?

1. Regenerate the full image for every small edit instead of using the supported editing input.
2. Use a preview capability for a workload whose production policy prohibits previews.
3. Assume prompt wording guarantees exact pixel placement without validating the output.
4. Moderate the generated response before it is returned or acted on when transparency is required.
5. Discard the reference image even though subject consistency is required.

Correct Answer: D

Correct Answer

Answer D is correct because this choice directly implements the required decision for output format choice when transparency is required. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would regenerate the full image for every small edit instead of using the supported editing input. That can address a neighboring concern, but it does not implement output format choice when transparency is required; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement output format choice when transparency is required; the decisive requirement in the stem therefore remains unresolved.

Answer C is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement output format choice when transparency is required; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement output format choice when transparency is required; the decisive requirement in the stem therefore remains unresolved.

 

Question 20

A multimodal workload requires rejecting an unsupported control instead of assuming cross-model parity. Which implementation best preserves the required visual evidence and policy boundary?

1. Generate at an arbitrary aspect ratio and crop essential content afterward.
2. Assume prompt wording guarantees exact pixel placement without validating the output.
3. Reject or redesign the request when the selected model does not support the requested control.
4. Discard the reference image even though subject consistency is required.
5. Use a preview capability for a workload whose production policy prohibits previews.

Correct Answer: C

Correct Answer

Answer C is correct because this choice directly implements the required decision for rejecting an unsupported control instead of assuming cross-model parity. It addresses the scenario at the correct stage of the Azure AI solution.

Incorrect Answers

Answer A is incorrect because this choice would generate at an arbitrary aspect ratio and crop essential content afterward. That can address a neighboring concern, but it does not implement rejecting an unsupported control instead of assuming cross-model parity; the decisive requirement in the stem therefore remains unresolved.

Answer B is incorrect because this choice would assume prompt wording guarantees exact pixel placement without validating the output. That can address a neighboring concern, but it does not implement rejecting an unsupported control instead of assuming cross-model parity; the decisive requirement in the stem therefore remains unresolved.

Answer D is incorrect because this choice would discard the reference image even though subject consistency is required. That can address a neighboring concern, but it does not implement rejecting an unsupported control instead of assuming cross-model parity; the decisive requirement in the stem therefore remains unresolved.

Answer E is incorrect because this choice would use a preview capability for a workload whose production policy prohibits previews. That can address a neighboring concern, but it does not implement rejecting an unsupported control instead of assuming cross-model parity; the decisive requirement in the stem therefore remains unresolved.