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.

• ✗

  A billing unit roughly equal to one character in the input or output text

  Why it's wrong here

  This is incorrect because Azure OpenAI does not bill by character count. Tokenization splits text into subword units, and for English a token is roughly four characters, which equals about three-quarters of a word—not a single character. Since a character-level unit would be far smaller, pricing that way would drastically overstate token consumption. Azure OpenAI's tokenizer is optimized for computational and cost efficiency, making character-based billing a misleading description.

• ✓

  A billing unit roughly equal to ¾ of an English word, counting both input and output

  Why this is correct

  This is the correct definition. In Azure OpenAI, a token is approximately three-quarters of an English word, though the exact count varies by tokenizer and language. Both input text (prompt tokens) and output text (completion tokens) are independently counted and billed at their respective rates. For example, a GPT-4 prompt and its generated completion both contribute token usage, which is reported in the API response and used to calculate the cost. This is the fundamental unit of consumption for pay-as-you-go pricing.

• ✗

  A subscription-based pricing model where a fixed number of API calls are included monthly

  Why it's wrong here

  This is incorrect because Azure OpenAI's pricing is consumption-based, not subscription-based per API call. While you can purchase provisioned throughput for reserved capacity, the default pay-as-you-go model charges per token processed, not per call. A fixed monthly number of API calls would ignore the variable length of each request and response, so it would not align with token-based billing. This option conflates alternative billing models with the actual token pricing mechanism.

• ✗

  Authentication tokens required to secure API calls to Azure OpenAI

  Why it's wrong here

  This is incorrect because authentication tokens are security credentials, not billing units. Azure OpenAI uses Azure Active Directory tokens or API keys to authorize and authenticate API requests, ensuring only permitted clients access the service. These tokens have no relationship to the text tokens used for metering usage and cost. Mixing up authentication tokens with pricing tokens is a common semantic error, but they serve entirely different purposes in the platform.

Go deeper

Related to this question