1Z0-1127-25 Using OCI Generative AI Service — 40 Questions

Question 1

You are a cloud architect at a healthcare company that uses OCI Generative AI Service to analyze patient records and generate clinical summaries. The service is deployed in the Frankfurt region with a dedicated AI cluster. Recently, the compliance team flagged that some generated summaries contain hallucinated diagnoses not present in the source records. They demand immediate mitigation. The current setup uses the default model (cohere.command-r-08-2024) with temperature=0.7, top_p=0.9, and max_tokens=2048. The application sends the entire patient record as a single prompt. You have access to OCI Logging, monitoring metrics (latency, request count, token count, safety filter rejections), and the AI service's model fine-tuning capability. You must reduce hallucinations while minimizing latency increase. What is the most effective course of action?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Reduce temperature to 0.2, top_p to 0.5, and fine-tune the model on a curated dataset of 5,000 clinical summaries with a learning rate of 0.00005 and batch size of 8.. Reducing temperature and top_p makes the model more deterministic, directly reducing the likelihood of hallucinated content. Fine-tuning on a curated dataset of 5,000 clinical summaries teaches the model domain-specific patterns and constraints, further minimizing hallucinations. This approach addresses the root cause without significantly increasing latency, as fine-tuning does not affect inference speed and lower sampling parameters add no computational overhead.

Answer

Switch to cohere.command-light model for faster inference and add a post-processing step using a BERT-based NER model to validate entities.

Answer

Increase max_tokens to 4096 and use chunked processing with overlapping context windows to provide more context.

Answer

Enable the safety filter with strict content moderation and set up OCI Logging to audit all generations.

Question 2

A developer wants to use the OCI Generative AI service to generate text using a Cohere model. Which SDK class should be used for inference calls?

Accepted Answer

GenerativeAiInferenceClient. The correct SDK class for inference calls with Cohere models in OCI Generative AI is `GenerativeAiInferenceClient`. This class is specifically designed to handle inference operations such as text generation, summarization, and chat completions, and it supports both Cohere and Meta Llama models. It provides methods like `generate_text`, `summarize_text`, and `chat` that directly interact with the inference endpoints.

Answer

CustomModelClient

Answer

AiInferenceClient

Answer

GenerativeAiClient

Question 3

A user has attached an IAM policy granting access to the generative-ai-family resource type, but API calls to the Generative AI service return a 403 Forbidden error. What is the most likely cause?

Accepted Answer

The policy does not specify the correct compartment or resource.. IAM policies in OCI are evaluated based on the resource type, the verb, and the compartment. Even if a policy grants access to the 'generative-ai-family' resource type, if the policy does not specify the correct compartment (or uses a condition that does not match the resource's compartment), the API call will be denied with a 403 Forbidden. The user must ensure the policy's 'allow' statement includes the appropriate compartment or uses a dynamic group or condition that matches the resource's location.

Answer

The service is not enabled in the tenancy.

Answer

The policy does not include the 'use' verb for the resource type.

Answer

The user has not created a Dedicated AI Cluster.

Question 4

After fine-tuning a Cohere Command model on a dataset of customer emails, the model performs well on validation data but poorly on new, unseen emails. Which action is most likely to improve generalization?

Accepted Answer

Expand the training dataset with more diverse examples.. The model is overfitting to the training data, which is a common issue when the dataset lacks diversity. Expanding the training dataset with more diverse examples exposes the model to a wider range of patterns and variations, reducing overfitting and improving generalization to unseen customer emails. In the context of Cohere Command models, this aligns with best practices for fine-tuning on OCI Generative AI Service, where data quality and diversity are critical for robust performance.

Answer

Increase the number of fine-tuning epochs.

Answer

Reduce the number of layers being fine-tuned.

Answer

Switch to a smaller model variant such as Cohere Light.

Question 5

When using OCI Generative AI with a fine-tuned model, what is the primary benefit of creating a dedicated AI cluster?

Accepted Answer

Consistent low latency and high throughput for production workloads.. A dedicated AI cluster in OCI Generative AI provides reserved compute capacity, ensuring consistent low latency and high throughput for production workloads. Unlike on-demand or auto-scaling setups, a dedicated cluster avoids resource contention with other tenants, making it ideal for latency-sensitive inference tasks with fine-tuned models.

Answer

Automatic scaling based on demand.

Answer

Reduced cost per inference token compared to on-demand.

Answer

Enhanced security through network isolation from other tenants.

Question 6

A financial services firm needs to extract named entities from legal contracts using OCI Generative AI. They require high accuracy and must handle domain-specific terminology. Which approach is most effective?

Accepted Answer

Fine-tune a Cohere Command model using a dataset of annotated legal contracts.. Fine-tuning a Cohere Command model on a domain-specific dataset of annotated legal contracts allows the model to learn the unique terminology, document structure, and entity boundaries required for high-accuracy extraction. This supervised approach adapts the model's weights to the target task, significantly outperforming zero-shot or prompt-based methods for specialized domains like legal contracts.

Answer

Use the base Cohere Command model with zero-shot prompting for entity extraction.

Answer

Use the Cohere Chat model with system prompts describing the entities.

Answer

Use the Cohere Embed model to generate embeddings and then train a separate classifier.

Question 7

Which TWO operations are supported by the OCI Generative AI inference API?

Accepted Answer

EmbedText. The OCI Generative AI inference API directly supports EmbedText for generating vector embeddings from input text, and GenerateText for producing natural language completions. These are the two core operations exposed by the service for embedding and text generation tasks.

Answer

SummarizeText

Answer

TranslateText

Answer

ChatCompletion

Question 8

A data scientist fine-tuned a model on OCI Gen AI using a dedicated AI cluster. After deployment, the model gives inaccurate results. Which troubleshooting step should they take first?

Accepted Answer

Check the training data for bias or quality issues. The most common cause of inaccurate results after fine-tuning is poor training data quality or bias. Before adjusting infrastructure or switching models, you must validate the dataset for issues like label errors, class imbalance, or sampling bias, as these directly degrade model performance regardless of compute resources.

Answer

Switch to a different base model

Answer

Increase the cluster size

Answer

Use a serverless endpoint

Question 9

A company wants to use OCI Generative AI but must comply with GDPR. Which feature ensures data residency?

Accepted Answer

Data localization with dedicated AI clusters. OCI Generative AI's data localization feature with dedicated AI clusters ensures that customer data remains within a specific geographic region, directly addressing GDPR's data residency requirements. By provisioning dedicated clusters in a chosen OCI region, the company can guarantee that data processing and storage do not cross borders without explicit consent, meeting the GDPR principle of restricting data transfer outside the EU/EEA.

Answer

Encryption at rest

Answer

Access control policies

Answer

Audit logging

Question 10

A user gets a 'Model not found' error when calling an OCI Gen AI endpoint. What is the most likely cause?

Accepted Answer

The endpoint is not deployed. The 'Model not found' error in OCI Generative AI typically occurs when the model endpoint has not been deployed or activated in the user's tenancy. Even if the model is available in the region and the request format is correct, the endpoint must be explicitly deployed (e.g., via the OCI Console, CLI, or SDK) before inference calls can succeed. This is a common prerequisite for using dedicated AI endpoints in OCI.

Answer

The model is not available in the region

Answer

The request format is wrong

Answer

The API key is invalid

Question 11

Which TWO configurations are required to use a custom fine-tuned model on OCI Gen AI?

Accepted Answer

A dedicated AI cluster. A dedicated AI cluster is required to host a custom fine-tuned model on OCI Generative AI because fine-tuned models are deployed on dedicated, single-tenant infrastructure that provides guaranteed compute capacity and isolation. Without a dedicated AI cluster, the model cannot be served for inference, as OCI Gen AI does not support serverless or shared endpoints for custom models.

Answer

A security list

Answer

An API key

Answer

A serverless endpoint

Question 12

Which THREE factors should be considered when choosing a base model for OCI Gen AI?

Accepted Answer

Availability in the desired region. OCI Gen AI models are deployed in specific regions, and availability varies by region due to data residency requirements and infrastructure placement. Before selecting a base model, you must verify that the model is available in your desired OCI region to avoid deployment failures or latency issues.

Answer

Supported languages

Answer

Training data format

Question 13

Which TWO steps are necessary to deploy a fine-tuned model on a dedicated AI cluster?

Accepted Answer

Create the dedicated AI cluster. Creating the dedicated AI cluster is the foundational step because it provisions the GPU-backed infrastructure required to host and run the fine-tuned model. Without this cluster, there is no compute environment to deploy the model onto, making it a prerequisite for any deployment action.

Answer

Set up a VCN with public subnet

Answer

Upload the model artifacts to OCI Object Storage

Answer

Obtain a third-party license

Question 14

Refer to the exhibit. A user runs this command and gets an error: 'InvalidParameter: unit-shape 'GPU.1' is not supported in this region. Supported shapes: GPU.2, GPU.4'. What should the user do?

Accepted Answer

Use unit-shape GPU.2. The error message explicitly states that 'GPU.1' is not supported in the current region and lists 'GPU.2' and 'GPU.4' as supported shapes. Option A correctly instructs the user to use a supported shape, such as GPU.2, which resolves the invalid parameter error without changing other configurations.

Answer

Increase unit count

Answer

Change compartment

Answer

Use a different region

Question 15

Users report that inference requests to the OCI Generative AI service are taking longer than expected. The application uses the on-demand endpoint. What is the most likely cause of the increased latency?

Accepted Answer

The on-demand endpoint experiences shared resource contention.. The on-demand endpoint in OCI Generative AI uses a shared infrastructure pool where compute resources are allocated dynamically. When multiple users or applications send concurrent inference requests, resource contention occurs, leading to queuing delays and increased latency. This is the most likely cause because the question specifies that users are experiencing slower-than-expected responses, and shared resource contention is a known behavior of on-demand endpoints.

Answer

The inference model is not fine-tuned for the use case.

Answer

The selected model is too large for the use case.

Answer

The API request timeout is set too low.

Question 16

A company is deploying a customer-facing chatbot using OCI Generative AI. They need to prevent the model from generating offensive or harmful content. Which feature should they implement?

Accepted Answer

Enable OCI Generative AI guardrails with content filtering.. OCI Generative AI guardrails provide built-in content filtering that automatically detects and blocks offensive or harmful content in both user inputs and model outputs. This is the recommended, native feature for safety compliance, eliminating the need for custom post-processing and ensuring consistent policy enforcement across all chatbot interactions.

Answer

Custom post-processing to scan each response.

Answer

Limit the user input length.

Answer

Use a smaller model that is less capable of generating harm.

Question 17

Which TWO are benefits of using dedicated AI clusters for OCI Generative AI?

Accepted Answer

Guaranteed throughput. Dedicated AI clusters provide guaranteed throughput because the compute and networking resources are exclusively allocated to your workloads, eliminating contention from other tenants. This ensures that the model inference capacity is always available at the level you provision, which is critical for production applications with strict performance requirements.

Answer

Automatic model updates

Answer

Lower cost than on-demand for all workloads

Answer

No need to manage scaling

Question 18

Refer to the exhibit. The user requested a long story but the response is cut short. What is the most likely cause?

Accepted Answer

The max_tokens parameter is too low for the requested length.. The `max_tokens` parameter explicitly limits the number of tokens (words/subwords) the model can generate in a single response. If the user requests a long story but `max_tokens` is set too low, the model will stop generating once the token limit is reached, resulting in a truncated output. This is a common configuration issue when using OCI Generative AI Service or any large language model API.

Answer

The model content filter blocked part of the output.

Answer

There was a network error during inference.

Answer

The model is not capable of generating long stories.

Question 19

Refer to the exhibit. The group DataScientists can run inference but cannot fine-tune a model on a dedicated AI cluster. Which additional policy statement is required to allow fine-tuning?

Accepted Answer

Allow group DataScientists to manage dedicated-ai-clusters in compartment ABC. To fine-tune a model on a dedicated AI cluster, the user must have the `manage` permission on the `dedicated-ai-clusters` resource type in the compartment. The `inspect` permission (Option A) only allows viewing metadata, not performing write operations like fine-tuning. The `use` permission (Option D) is insufficient for management tasks. Option C is invalid because `generative-ai-fine-tune` is not a valid resource type in OCI IAM policies.

Answer

Allow group DataScientists to inspect dedicated-ai-clusters in compartment ABC

Answer

Allow group DataScientists to use generative-ai-fine-tune in compartment ABC

Answer

Allow group DataScientists to use dedicated-ai-clusters in compartment ABC

Question 20

Refer to the exhibit. In this RAG pipeline, what is the role of the 'embedding_model' variable?

Accepted Answer

It converts text into vector representations for similarity search.. In a Retrieval-Augmented Generation (RAG) pipeline, the 'embedding_model' variable is responsible for converting input text (such as user queries or document chunks) into dense vector representations. These vectors are then used to perform similarity search against a vector database, enabling the retrieval of the most relevant context for the generative model. This is a core function of the embedding step in RAG, not a guardrail, fine-tuning, or text generation role.

Answer

It applies guardrails to filter content.

Answer

It fine-tunes the model on the provided texts.

Answer

It generates text completions based on prompts.