AI-300 Practice Question 97: Deployment Name Not Found
Microsoft Machine Learning Operations Engineer Associate (AI-300) · Test 10 · question 7 of 10
A team deployed the gpt-4.1 model under the deployment name support-chat. Their application code passes model="gpt-4.1" in each request and receives a 404 DeploymentNotFound error. What should they change?
Show answer and explanation
Answer:
Requests to a Foundry resource identify a model by the deployment name that was chosen when the model was deployed, not by the underlying model name. Because no deployment is named gpt-4.1, the service returns a not-found error. Passing support-chat fixes it (option B).
- A is wrong because a rate limit problem returns 429, not 404, and the request never finds the deployment. - C is wrong because a missing role produces an authorization error such as 401 or 403, not a not-found error. - D is wrong because the error says that no deployment has that name, so the API version is not the cause.
Reference: Deploy models to Foundry Models