AI-900 Practice Question: Describe features of generative AI workloads on Azure

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  They refer to how many GPUs are used for model training

  Why it's wrong here

  GPU count is an infrastructure detail: more GPUs or TPUs accelerate the arithmetic needed for training and inference, but they do not affect how many examples are placed in the prompt. Zero-shot and few-shot describe how the task is specified to the model—whether the prompt contains zero or multiple demonstrations—not the hardware resources used to run it. The same model can process zero-shot or few-shot prompts on a single GPU or a large cluster, so shot count and GPU count are unrelated concepts.

• ✓

  Zero-shot uses no examples; few-shot provides multiple examples in the prompt to guide responses

  Why this is correct

  Zero-shot prompting means providing the model with only a text instruction or question and no illustrative examples, forcing it to rely on its pretrained knowledge to produce a relevant response. Few-shot prompting includes multiple examples in the prompt that demonstrate the desired input-to-output mapping, so the model follows the pattern to generate consistently formatted or reasoned output. In both cases the model's weights stay fixed; the examples only condition its next-token predictions at inference time rather than representing any form of training.

• ✗

  They refer to how many training epochs the model underwent

  Why it's wrong here

  An epoch is a single complete pass through the entire training dataset during model optimization, when weights are updated via backpropagation. Zero-shot and few-shot, by contrast, describe inference-time prompting: they count how many example question-answer pairs or demonstrations are placed in the prompt before the model generates output. The model is already trained before any shot count is relevant, and adding examples to a prompt does not trigger additional training epochs.

• ✗

  Zero-shot is for beginners; few-shot is for experts

  Why it's wrong here

  Zero-shot and few-shot are not learner-expert labels; the shot count is entirely about the prompt's composition, not the user's skill level. A beginner can easily write a few-shot prompt containing several examples, and an advanced user frequently chooses zero-shot prompts for quick, low-cost interactions. The terms simply indicate how many demonstrations appear in the request—zero or multiple—and have no connection to whether the person crafting the prompt is experienced or new.

Go deeper

Related to this question