GES-C01 無料問題集「Snowflake SnowPro® Specialty: Gen AI Certification」

（A）Enabling Azure OpenAI models via the ' ENABLE_CORTEX_ANALYST_MODEL_AZURE_OPENAF parameter ensures that all customer data and prompts remain within Snowflake's governance boundary and fully respect RBAC policies for those models.

（B）Snowflake's view provides granular usage information for REST API requests to cortex Analyst, including tokens processed per model.

（C）Cortex Analyst credit usage is based on the number of messages processed, at a rate of 67 Credits per 1 ,000 messages, and only successful responses (HTTP 200) are counted.

（D）Credit consumption for Cortex Analyst is primarily based on the number of tokens processed by the underlying LLMs, with more complex natural language questions leading to higher token usage and costs.

（E）The view can be queried to track detailed information about cortex Analyst requests, including generated SQL and any errors.

（A）Selecting a 'LARGE' or 'X-LARGE' standard warehouse will significantly improve the processing speed and reduce the per-page cost for AI_PARSE_DOCUMENT.

（B）If a document is a HTML file, the billing is based on processing each chunk of 3,000 characters as a page.

（C）The primary billing metric for in LAYOUT mode is the number of tokens processed, which includes both input and output tokens.

（D）To reduce costs, the team should aim to process documents in batches, ensuring each query processes a maximum of 1 ,000 documents to avoid errors.

（E）For optimal results with multilingual documents, the LAYOUT mode supports a broader range of languages than OCR mode.

（A）The query latency for cross-region inference with AI_COMPLETE is consistently low and predictable, as Snowflake's architecture is designed to completely negate the impact of geographical distance and network variations.

（B）To enable cross-region inference for the US East (Ohio) account, the administrator would execute the command: ALTER ACCOUNT SET AWS_US' ; to allow inference requests to be processed in any AWS US region where the model is available. CORTEX_ENABLED_CROSS REGION =AWS_US' ;

（C）The llama4-maverick model is listed as natively available in AWS US East 1 (N. Virginia) and is supported for cross-region inference (AWS US Cross-Region), validating it as a suitable target for inference from US East (Ohio).

（D）Cross-region inference is fully supported for AI_COMPLETE in U.S. SnowGov regions for both inbound and outbound inference requests, provided the target model is available in the respective SnowGov region.

（E）User inputs, service-generated prompts, and the generated outputs from cross-region AI_COMPLETE calls are automatically stored or cached in the remote processing region to optimize performance for subsequent identical requests.

（A）The array contains more than 100 unique categories, exceeding the maximum allowed limit for the function.

（B）Snowflake Cortex functions, including 'CLASSIFY_TEXT , currently do not support integration with dynamic tables within data pipelines.

（C）The input text being processed by 'CLASSIFY _ TEXT includes extensive non-plain English content, such as code blocks, which causes the function to fail with an error.

（D）The role used by the data engineer, despite having 'SNOWFLAKE.CORTEX_USER, lacks the fundamental 'USAGE privilege on the database where the text data is stored.

（E）The 'task_description' provided in the optional arguments for 'CLASSIFY_TEXT exceeds the recommended length of approximately 50 words, leading to a validation error.

（A）Option E

（B）Option D

（C）Option B

（D）Option C

（E）Option A

（A）Option E

（B）Option D

（C）Option B

（D）Option C

（E）Option A

（A）Including a

（B）Setting

（C）For a multi-turn conversation, previous user prompts and model responses should be passed in the

（D）The

（E）The

（A）Cortex Guard is inherently part of all Cortex LLM functions and does not require explicit enablement via 'guardrails: TRUE for SQL functions or the REST API.

（B）The underlying model for Cortex Guard is Meta's Llama Guard 3, and its processing costs are separate from the primary LLM inference costs.

（C）Cortex Guard operates by evaluating responses after the LLM has fully generated its content, and it incurs additional compute cost for both input and output tokens during its processing.

（D）Cortex Guard can be configured with a custom message using the argument in the options object for both 'COMPLETE SQL function and the Cortex REST API.

（E）When Cortex Guard is enabled and a response is deemed unsafe, the LLM-generated output is replaced with a predefined message, and only the input tokens for Cortex Guard processing ('guard_tokens') are billed, not the potentially unsafe completion tokens.

（A）if a If a model specified in 'TRY_COMPLETE is not allowed by the and no RBAC grant exists for it, the 'TRY_COMPLETE function will automatically return 'NULL' to prevent a hard error.

（B）The 'TRY_COMPLETE function can return a 'NULL ' value if the underlying LLM experiences a temporary internal processing error during response generation, allowing the query execution to continue without halting.

（C）To provide a stateful conversational experience with 'TRY_COMPLETE, the function automatically retains context from previous calls within the same session.

（D）The parameter primarily governs access for models invoked by their plain names with but 'TRY COMPLETE can still access models if the user has been granted appropriate RBAC on a fully-qualified model object.

（E）If 'TRY_COMPLETE fails to perform its operation and returns ' NULL ' , the tokens that would have been processed for that specific call are not billed, ensuring cost efficiency for failed operations.

（A）The cost for fine-tuning is incurred based on the number of tokens used in training, calculated as:

（B）The fine-tuned model will automatically appear in the Snowflake Model Registry and is available for sharing with other accounts using Data Sharing, even if it contains user code.

（C）To create the fine-tuning job, the engineer should use the SQL command:

（D）The training data query result must explicitly include columns named

（E）The 'max_epochs' option in the 'options' JSON object can be set to any positive integer to control the training duration, but it is capped at a maximum of 50 epochs.

（A）Option E

（B）Option D

（C）Option B

（D）Option C

（E）Option A

（A）Option E

（B）Option D

（C）Option B

（D）Option C

（E）Option A

（A）Option E

（B）Option D

（C）Option B

（D）Option C

（E）Option A

（A）To optimize for latency, Cortex Analyst recommends always using the smallest possible LLM for the summarization agent, such as Llama 3.1 8B, even if it has a slightly higher error rate in rewriting questions.

（B）The summarization agent in Cortex Analyst is primarily responsible for generating SQL queries from conversation history, thus requiring an LLM optimized for text- to-SQL tasks.

（C）Multi-turn conversation support primarily relies on caching previous SQL query results and re-executing them for follow-up questions, avoiding additional LLM calls for context summarization.

（D）Cortex Analyst directly passes the entire raw conversation history to every LLM call for all agents to ensure full context, which generally improves performance.

（E）A dedicated LLM summarization agent is introduced before the original workflow to distill conversation history into a concise context for subsequent agents, with Llama 3.1 70B identified as a suitable model due to its high summarization quality.