Practice Free NCP-AI Exam Online Questions
How should a non-text-generation LLM endpoint be tested?
- A . Copy the Sample API Code and URL and run it in the application.
- B . Select Custom Request and upload the provided test file, and click Test.
- C . Create a Sample Request, choose a default prompt, and click Test.
- D . Open the URL on a Browser on the HTTPS web page.
An administrator is managing a Nutanix AI cluster used for NLP (Natural Language Processing) training. A data scientist reports that training jobs intermittently stall and fail to complete within the expected time window. The administrator reviews the performance data for the VM hosting the job and finds:
The VM is configured with passthrough access to a dedicated GPU
Memory ballooning is active, and swap usage is increasing
CPU utilization is moderate (~60%)
GPU utilization is stable and high (~85%)
The VM has 8 vCPUs and 24 GB of RAM assigned
NCC shows no hardware or driver issues
What is the most appropriate optimization to improve workload stability and performance?
- A . Reduce the number of vCPUs allocated to lower the CPU scheduling overhead.
- B . Add additional vGPUs to the VM to reduce processing time.
- C . Increase the VM’s RAM to eliminate memory ballooning and swap usage.
- D . Disable GPU passthrough and use a shared vGPU profile instead.
An administrator is configuring a container registry to store custom models for secure Nutanix Enterprise AI deployment. During testing, the administrator notices the registry is accessible both via HTTP and HTTPS, with a self-signed certificate.
Which two actions should the administrator take to meet the deployment prerequisites? (Choose two.)
- A . Disable HTTP on the container registry.
- B . Install Nutanix Enterprise AI in offline mode.
- C . Replace the certificate with a trusted TLS/SSL certificate.
- D . Ensure registry runs on the same node as the cluster.