{"id":19384,"date":"2026-09-23T05:31:30","date_gmt":"2026-09-23T05:31:30","guid":{"rendered":"https:\/\/www.examlabs.com\/certification\/?p=19384"},"modified":"2026-09-23T05:31:30","modified_gmt":"2026-09-23T05:31:30","slug":"snowflake-snowpro-specialty-gen-ai-ges-c01-practice-test-questions-and-exam-dumps-part-1-q1-20","status":"publish","type":"post","link":"https:\/\/www.examlabs.com\/certification\/snowflake-snowpro-specialty-gen-ai-ges-c01-practice-test-questions-and-exam-dumps-part-1-q1-20\/","title":{"rendered":"Snowflake SnowPro Specialty Gen AI GES-C01 Practice Test Questions and Exam Dumps Part 1 Q1-20"},"content":{"rendered":"

View Full <\/b>Snowflake SnowPro Specialty Gen AI GES-C01<\/b> Exam Dumps<\/b><\/a> and Practice Test Dumps<\/b><\/p>\n

 <\/p>\n

Question 1. Which Snowflake capability provides access to generative AI and machine learning functionality directly within the Snowflake platform?<\/b><\/p>\n

    \n
  1. <\/b> Snowflake Cortex<\/span><\/li>\n
  2. <\/b> Snowflake Tasks<\/span><\/li>\n
  3. <\/b> Snowflake Streams<\/span><\/li>\n
  4. <\/b> Snowflake Stages<\/span><\/li>\n<\/ol>\n

    Correct Answer: 1. Snowflake Cortex<\/b><\/p>\n

    Explanation:<\/b>
    \n<\/b>Snowflake Cortex provides AI and machine learning capabilities directly within the Snowflake environment. It enables users and applications to work with large language models, embeddings, search, and other AI functionality while keeping data workflows closely integrated with Snowflake. Cortex can support use cases such as text generation, summarization, classification, semantic search, and conversational analytics. By providing these capabilities within the data platform, organizations can reduce the need to move data to separate AI environments for many workloads. The specific Cortex functions and services available depend on the use case and Snowflake capabilities being used.<\/span><\/p>\n

    Question 2. Which Snowflake function is designed to generate vector embeddings from text?<\/b><\/p>\n

      \n
    1. <\/b> AI_SUMMARIZE<\/span><\/li>\n
    2. <\/b> EMBED_TEXT<\/span><\/li>\n
    3. <\/b> CLASSIFY_TEXT<\/span><\/li>\n
    4. <\/b> COMPLETE<\/span><\/li>\n<\/ol>\n

      Correct Answer: 2. EMBED_TEXT<\/b><\/p>\n

      Explanation:<\/b>
      \n<\/b>EMBED_TEXT is used to generate vector embeddings that represent the semantic meaning of text. These numerical vectors can then be used for similarity comparisons, semantic search, recommendation systems, and retrieval-augmented generation workflows. Instead of relying only on exact keyword matching, embeddings allow content to be represented according to meaning and relationships between concepts. The resulting vectors can be stored and searched using appropriate vector capabilities. In a RAG architecture, documents and user queries can both be converted into embeddings, allowing the system to identify document chunks that are semantically relevant to the user’s request.<\/span><\/p>\n

      Question 3. What is the primary purpose of Retrieval-Augmented Generation (RAG)?<\/b><\/p>\n

        \n
      1. <\/b> To replace all enterprise databases with an LLM<\/span><\/li>\n
      2. <\/b> To train an LLM from scratch using every available document<\/span><\/li>\n
      3. <\/b> To retrieve relevant enterprise information and provide it as context to an LLM<\/span><\/li>\n
      4. <\/b> To increase the number of parameters inside an existing LLM<\/span><\/li>\n<\/ol>\n

        Correct Answer: 3. To retrieve relevant enterprise information and provide it as context to an LLM<\/b><\/p>\n

        Explanation:<\/b>
        \n<\/b>Retrieval-Augmented Generation combines information retrieval with language-model generation. When a user submits a query, the retrieval component searches a relevant knowledge source and identifies useful information, often from documents or other enterprise data. That information is then supplied to the language model as context before the model generates its response. This approach can help responses reflect organization-specific or current information without requiring the model itself to be retrained for every data change. Effective RAG systems depend on appropriate document preparation, chunking, embedding or search strategies, retrieval quality, and prompt construction.<\/span><\/p>\n

        Question 4. Which Snowflake Cortex function is primarily associated with generating text completions using a language model?<\/b><\/p>\n

          \n
        1. <\/b> CLASSIFY_TEXT<\/span><\/li>\n
        2. <\/b> COMPLETE<\/span><\/li>\n
        3. <\/b> EMBED_TEXT<\/span><\/li>\n
        4. <\/b> AI_FILTER<\/span><\/li>\n<\/ol>\n

          Correct Answer: 2. COMPLETE<\/b><\/p>\n

          Explanation:<\/b>
          \n<\/b>The COMPLETE function is associated with generating text using a language model. It can be used in applications where an input prompt or context is provided and the model generates a corresponding response. Common generative AI scenarios include drafting text, answering questions, transforming content, and producing other language-based outputs. The quality of the result depends on factors such as the model selected, prompt design, supplied context, and input data. COMPLETE differs from embedding functions because embeddings represent semantic information as vectors rather than generating natural-language text.<\/span><\/p>\n

          Question 5. What is the main purpose of vector embeddings in a generative AI application?<\/b><\/p>\n

            \n
          1. <\/b> To store passwords securely<\/span><\/li>\n
          2. <\/b> To compress database tables<\/span><\/li>\n
          3. <\/b> To increase network bandwidth<\/span><\/li>\n
          4. <\/b> To represent content as numerical vectors capturing semantic relationships<\/span><\/li>\n<\/ol>\n

            Correct Answer: 4. To represent content as numerical vectors capturing semantic relationships<\/b><\/p>\n

            Explanation:<\/b>
            \n<\/b>Vector embeddings represent text or other supported content as numerical vectors that capture aspects of semantic meaning. Content with related meanings can produce vectors that are closer together in a suitable vector space. This makes embeddings useful for semantic search, document retrieval, recommendations, clustering, and RAG systems. For example, a user query can be converted into an embedding and compared with document embeddings to identify relevant content even when the wording is different. Embeddings do not directly generate natural-language responses; instead, they provide a machine-readable representation that can support retrieval and other AI workflows.<\/span><\/p>\n

            Question 6. Which Snowflake capability is designed to support intelligent search across enterprise content?<\/b><\/p>\n

              \n
            1. <\/b> Cortex Search<\/span><\/li>\n
            2. <\/b> Snowflake Tasks<\/span><\/li>\n
            3. <\/b> Snowflake Streams<\/span><\/li>\n
            4. <\/b> Snowflake Time Travel<\/span><\/li>\n<\/ol>\n

              Correct Answer: 1. Cortex Search<\/b><\/p>\n

              Explanation:<\/b>
              \n<\/b>Cortex Search is designed to provide intelligent search capabilities over enterprise content. It can support retrieval scenarios where users need relevant information from organizational data rather than relying exclusively on exact keyword matches. Search functionality is particularly useful in RAG applications because the retrieved information can be supplied to a language model as contextual grounding. Effective search depends on the quality and structure of the indexed content and the configuration of the search solution. Cortex Search is therefore relevant to applications such as enterprise knowledge assistants, document discovery, and question-answering systems.<\/span><\/p>\n

              Question 7. Which Snowflake account parameter is associated with controlling access to Cortex AI models through an allowlist?<\/b><\/p>\n

                \n
              1. <\/b> ENABLE_AI_FEATURES<\/span><\/li>\n
              2. <\/b> CORTEX_MODELS_ALLOWLIST<\/span><\/li>\n
              3. <\/b> AI_MODEL_ACCESS<\/span><\/li>\n
              4. <\/b> CORTEX_MODEL_SECURITY<\/span><\/li>\n<\/ol>\n

                Correct Answer: 2. CORTEX_MODELS_ALLOWLIST<\/b><\/p>\n

                Explanation:<\/b>
                \n<\/b>CORTEX_MODELS_ALLOWLIST is associated with controlling which Cortex models can be accessed through an allowlist configuration. Governance and administrative controls are important when organizations make generative AI functionality available to users because administrators may need to determine which models can be used. An allowlist approach can help organizations manage approved model access according to internal requirements. Model-access controls should be considered alongside account roles, privileges, data-access permissions, and organizational governance policies. These controls help separate the technical availability of AI capabilities from the authorization of individual users or workloads.<\/span><\/p>\n

                Question 8. In a RAG workflow, what is the main role of the retrieval stage?<\/b><\/p>\n

                  \n
                1. <\/b> Generate the final answer without using external information<\/span><\/li>\n
                2. <\/b> Permanently retrain the language model<\/span><\/li>\n
                3. <\/b> Retrieve relevant document chunks based on the user’s query<\/span><\/li>\n
                4. <\/b> Remove all context from the prompt<\/span><\/li>\n<\/ol>\n

                  Correct Answer: 3. Retrieve relevant document chunks based on the user’s query<\/b><\/p>\n

                  Explanation:<\/b>
                  \n<\/b>The retrieval stage of a RAG workflow identifies information that is relevant to the user’s query. Source material is commonly divided into manageable chunks and represented in a searchable form. When a query arrives, the retrieval component searches those sources and returns the most relevant chunks or passages. The retrieved information can then be incorporated into the prompt supplied to the language model. Retrieval quality has a direct effect on the usefulness of the final response because irrelevant or incomplete context can lead to weaker answers. Retrieval therefore acts as the bridge between enterprise information and the generation stage.<\/span><\/p>\n

                  Question 9. What is the primary purpose of prompt engineering in a generative AI application?<\/b><\/p>\n

                    \n
                  1. <\/b> To design instructions and context that guide the model toward the desired output<\/span><\/li>\n
                  2. <\/b> To physically increase the model’s parameter count<\/span><\/li>\n
                  3. <\/b> To replace all data governance controls<\/span><\/li>\n
                  4. <\/b> To eliminate the need for an inference model<\/span><\/li>\n<\/ol>\n

                    Correct Answer: 1. To design instructions and context that guide the model toward the desired output<\/b><\/p>\n

                    Explanation:<\/b>
                    \n<\/b>Prompt engineering involves designing instructions, context, constraints, examples, and formatting requirements that guide a generative AI model toward a desired result. A well-designed prompt can clarify the task, establish relevant context, specify the expected response format, and reduce ambiguity. In RAG applications, retrieved information can be incorporated into the prompt so the model has access to relevant source material. Prompt engineering does not modify the underlying model parameters. Instead, it focuses on how information and instructions are presented to the model during inference. Effective prompts should be tested against representative inputs to evaluate consistency and quality.<\/span><\/p>\n

                    Question 10. Which approach can help ensure that a generative AI model returns information in a predictable structure?<\/b><\/p>\n

                      \n
                    1. <\/b> Structured output or response-format configuration<\/span><\/li>\n
                    2. <\/b> Removing all instructions from the prompt<\/span><\/li>\n
                    3. <\/b> Increasing database storage capacity<\/span><\/li>\n
                    4. <\/b> Disabling retrieval<\/span><\/li>\n<\/ol>\n

                      Correct Answer: 1. Structured output or response-format configuration<\/b><\/p>\n

                      Explanation:<\/b>
                      \n<\/b>Structured output or response-format configuration can help constrain a generative AI response to a predictable structure. This is useful when an application needs to process model output programmatically, such as extracting fields, creating records, or passing generated information to another component. Clearly specifying the expected schema or format can reduce ambiguity and make downstream processing more reliable. However, structured output does not automatically guarantee that every generated value is factually correct. Applications should still validate important outputs and apply appropriate business rules, security controls, and error handling before using generated data in consequential workflows.<\/span><\/p>\n

                      Question 11. Which Snowflake capability is intended to extract useful information from documents and other unstructured content?<\/b><\/p>\n

                        \n
                      1. <\/b> Snowflake Tasks<\/span><\/li>\n
                      2. <\/b> Document AI<\/span><\/li>\n
                      3. <\/b> Snowflake Streams<\/span><\/li>\n
                      4. <\/b> Dynamic Tables<\/span><\/li>\n<\/ol>\n

                        Correct Answer: 2. Document AI<\/b><\/p>\n

                        Explanation:<\/b>
                        \n<\/b>Document AI is designed to help organizations process and extract information from documents and other unstructured content. This capability is useful when important business information exists in files rather than in traditionally structured relational tables. Extracted information can potentially be incorporated into downstream data processing and AI workflows. For example, document processing can help identify fields or content that can later be searched, analyzed, or used as context for generative AI applications. Document AI addresses document understanding and extraction rather than serving as a general-purpose database management or scheduling feature.<\/span><\/p>\n

                        Question 12. Why can keeping AI workloads close to enterprise data be beneficial?<\/b><\/p>\n

                          \n
                        1. <\/b> It guarantees that every model response is correct<\/span><\/li>\n
                        2. <\/b> It eliminates the need for data governance<\/span><\/li>\n
                        3. <\/b> It can reduce unnecessary data movement and simplify governance<\/span><\/li>\n
                        4. <\/b> It makes every AI model open source<\/span><\/li>\n<\/ol>\n

                          Correct Answer: 3. It can reduce unnecessary data movement and simplify governance<\/b><\/p>\n

                          Explanation:<\/b>
                          \n<\/b>Keeping AI workloads close to enterprise data can reduce unnecessary data movement and make governance easier to manage. Moving sensitive or large datasets between multiple platforms can introduce additional integration, security, and operational considerations. When AI functionality is integrated with the data platform, organizations can often apply existing data-access controls and governance processes more directly. This does not guarantee that an AI application is secure or accurate; appropriate permissions, policies, monitoring, and validation are still required. The primary benefit is that data and AI processing can be brought into a more integrated architecture.<\/span><\/p>\n

                          Question 13. Which function is designed to classify text into categories?<\/b><\/p>\n

                            \n
                          1. <\/b> CLASSIFY_TEXT<\/span><\/li>\n
                          2. <\/b> COMPLETE<\/span><\/li>\n
                          3. <\/b> EMBED_TEXT<\/span><\/li>\n
                          4. <\/b> AI_SUMMARIZE<\/span><\/li>\n<\/ol>\n

                            Correct Answer: 1. CLASSIFY_TEXT<\/b><\/p>\n

                            Explanation:<\/b>
                            \n<\/b>CLASSIFY_TEXT is intended for text-classification use cases in which text needs to be assigned to predefined categories. Examples can include categorizing customer feedback, routing support messages, or assigning documents to business-defined classifications. Classification differs from text generation because the objective is to determine an appropriate category rather than produce an open-ended response. The quality of classification depends on the quality of the input, category definitions, and model behavior. Organizations should evaluate classification results against representative data and establish appropriate handling for ambiguous or unexpected inputs.<\/span><\/p>\n

                            Question 14. How does vector similarity search identify relevant content?<\/b><\/p>\n

                              \n
                            1. <\/b> By matching only identical words in two documents<\/span><\/li>\n
                            2. <\/b> By identifying text or objects whose vectors are mathematically similar to the query vector<\/span><\/li>\n
                            3. <\/b> By randomly selecting documents from a database<\/span><\/li>\n
                            4. <\/b> By comparing database passwords<\/span><\/li>\n<\/ol>\n

                              Correct Answer: 2. By identifying text or objects whose vectors are mathematically similar to the query vector<\/b><\/p>\n

                              Explanation:<\/b>
                              \n<\/b>Vector similarity search compares numerical representations of content to determine which items are mathematically close to a query representation. A query can be converted into an embedding vector, and that vector can then be compared with vectors representing stored documents or other content. Similarity metrics help identify content that is semantically related even when it does not contain the exact same words as the query. This makes vector search particularly useful for semantic retrieval and RAG applications. The quality of the results depends on factors such as the embedding model, source content, chunking strategy, and search configuration.<\/span><\/p>\n

                              Question 15. What is a primary purpose of Snowflake Cortex LLM functions?<\/b><\/p>\n

                                \n
                              1. <\/b> To replace Snowflake’s storage engine<\/span><\/li>\n
                              2. <\/b> To provide AI capabilities that can be invoked within Snowflake workflows<\/span><\/li>\n
                              3. <\/b> To disable access controls<\/span><\/li>\n
                              4. <\/b> To create physical network connections<\/span><\/li>\n<\/ol>\n

                                Correct Answer: 2. To provide AI capabilities that can be invoked within Snowflake workflows<\/b><\/p>\n

                                Explanation:<\/b>
                                \n<\/b>Snowflake Cortex LLM functions provide AI capabilities that can be invoked as part of Snowflake data and application workflows. Depending on the supported function, users can perform tasks such as text generation, classification, summarization, translation, or embedding generation. Integrating AI functionality with data workflows can reduce the need to move information into separate processing environments for every use case. These functions operate alongside Snowflake’s existing security and data-management capabilities, although organizations must still configure appropriate roles, privileges, model access, and governance controls for their workloads.<\/span><\/p>\n

                                Question 16. Which architecture is commonly used to provide enterprise-specific information to a generative AI model without retraining the model for every data update?<\/b><\/p>\n

                                  \n
                                1. <\/b> Retrieval-Augmented Generation (RAG)<\/span><\/li>\n
                                2. <\/b> Database replication only<\/span><\/li>\n
                                3. <\/b> Static HTML rendering<\/span><\/li>\n
                                4. <\/b> Network load balancing<\/span><\/li>\n<\/ol>\n

                                  Correct Answer: 1. Retrieval-Augmented Generation (RAG)<\/b><\/p>\n

                                  Explanation:<\/b>
                                  \n<\/b>Retrieval-Augmented Generation allows a language model to use relevant external information at inference time. Instead of retraining the model whenever enterprise information changes, the application retrieves current or organization-specific content and supplies it to the model as context. This architecture is particularly useful for knowledge bases, internal documentation, policies, and other information that may change frequently. A typical RAG pipeline includes source preparation, chunking, indexing or embedding, retrieval, prompt construction, and model generation. The approach does not eliminate the need for model evaluation or governance, but it can provide a practical way to ground responses in external information.<\/span><\/p>\n

                                  Question 17. Which Snowflake function can be used to create embeddings that support semantic retrieval?<\/b><\/p>\n

                                    \n
                                  1. <\/b> AI_SUMMARIZE<\/span><\/li>\n
                                  2. <\/b> CLASSIFY_TEXT<\/span><\/li>\n
                                  3. <\/b> EMBED_TEXT<\/span><\/li>\n
                                  4. <\/b> COMPLETE<\/span><\/li>\n<\/ol>\n

                                    Correct Answer: 3. EMBED_TEXT<\/b><\/p>\n

                                    Explanation:<\/b>
                                    \n<\/b>EMBED_TEXT can be used to create numerical vector representations of text for semantic retrieval and related AI applications. During an embedding workflow, source content is transformed into vectors that capture semantic characteristics. Queries can also be embedded so that the application can compare the query vector with stored content vectors. This enables retrieval based on semantic similarity rather than requiring exact textual matches. Embeddings are therefore an important component in many RAG and search architectures. The embedding model and the way source content is prepared can influence retrieval quality, so these components should be evaluated using representative application data.<\/span><\/p>\n

                                    Question 18. Which consideration is especially important when deploying generative AI with enterprise data?<\/b><\/p>\n

                                      \n
                                    1. <\/b> Governance, security, access control, and responsible AI practices<\/span><\/li>\n
                                    2. <\/b> Removing all user permissions<\/span><\/li>\n
                                    3. <\/b> Allowing unrestricted access to every dataset<\/span><\/li>\n
                                    4. <\/b> Ignoring model output quality<\/span><\/li>\n<\/ol>\n

                                      Correct Answer: 1. Governance, security, access control, and responsible AI practices<\/b><\/p>\n

                                      Explanation:<\/b>
                                      \n<\/b>Enterprise generative AI deployments require appropriate governance, security, access control, and responsible AI practices. Organizations should determine who can access particular datasets and AI capabilities, how sensitive information is handled, and how model outputs are evaluated and monitored. Access controls should be applied to data and AI resources according to organizational requirements. Responsible AI practices can also address issues such as hallucinations, inappropriate outputs, transparency, and human oversight. Generative AI functionality does not replace established security or governance processes. Instead, it introduces additional considerations that should be incorporated into the overall data and application architecture.<\/span><\/p>\n

                                      Question 19. Which Snowflake capability is designed to help users ask natural-language questions about structured enterprise data?<\/b><\/p>\n

                                        \n
                                      1. <\/b> Cortex Search<\/span><\/li>\n
                                      2. <\/b> Cortex Analyst<\/span><\/li>\n
                                      3. <\/b> Document AI<\/span><\/li>\n
                                      4. <\/b> Snowflake Streams<\/span><\/li>\n<\/ol>\n

                                        Correct Answer: 2. Cortex Analyst<\/b><\/p>\n

                                        Explanation:<\/b>
                                        \n<\/b>Cortex Analyst is designed to support natural-language interaction with structured enterprise data. It can help translate business-oriented questions into analytical queries by using semantic information about the underlying data. A semantic model can provide definitions, relationships, and business context that help the system interpret terms used by users. This makes the capability useful for conversational analytics where users may not know SQL or the exact structure of database tables. The quality of results depends on factors such as the underlying data, semantic definitions, permissions, and how well the business concepts are represented.<\/span><\/p>\n

                                        Question 20. Which sequence best represents a typical RAG workflow?<\/b><\/p>\n

                                          \n
                                        1. <\/b> Generate a response \u2192 delete source data \u2192 create embeddings \u2192 ask the user<\/span><\/li>\n
                                        2. <\/b> Train a new LLM \u2192 remove documents \u2192 generate random context \u2192 return output<\/span><\/li>\n
                                        3. <\/b> Split and prepare content \u2192 create embeddings or an index \u2192 retrieve relevant context \u2192 provide it to the LLM<\/span><\/li>\n
                                        4. <\/b> Disable retrieval \u2192 generate output \u2192 create database tables \u2192 remove the prompt<\/span><\/li>\n<\/ol>\n

                                          Correct Answer: 3. Split and prepare content \u2192 create embeddings or an index \u2192 retrieve relevant context \u2192 provide it to the LLM<\/b><\/p>\n

                                          Explanation:<\/b>
                                          \n<\/b>A typical RAG workflow begins by preparing source content, often by splitting documents into manageable chunks. Those chunks can then be indexed or converted into embeddings to support retrieval. When a user submits a query, the retrieval component identifies relevant information and returns suitable context. That context is incorporated into the prompt supplied to the language model, which then generates the response. The workflow allows the model to use external or enterprise-specific information during inference. Retrieval quality, chunking, indexing, prompt construction, and source-data quality all influence the usefulness and reliability of the final response.<\/span><\/p>\n

                                           <\/p>\n","protected":false},"excerpt":{"rendered":"

                                          View Full Snowflake SnowPro Specialty Gen AI GES-C01 Exam Dumps and Practice Test Dumps   Question 1. Which Snowflake capability provides access to generative AI and machine learning functionality directly within the Snowflake platform? Snowflake Cortex Snowflake Tasks Snowflake Streams Snowflake Stages Correct Answer: 1. Snowflake Cortex Explanation: Snowflake Cortex provides AI and machine learning […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[1648,1647],"tags":[],"_links":{"self":[{"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts\/19384"}],"collection":[{"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/comments?post=19384"}],"version-history":[{"count":1,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts\/19384\/revisions"}],"predecessor-version":[{"id":19385,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts\/19384\/revisions\/19385"}],"wp:attachment":[{"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/media?parent=19384"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/categories?post=19384"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/tags?post=19384"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}