To query your knowledgebase, submit the user question in a call to the QnA Maker Runtime API’s GenerateAnswer method as illustrated below. The GenerateAnswer method returns question/answer pairs that matched your question along with the source document name or URL that the answer was found in. Query against the test index. Exact qnaId to fetch from the knowledgebase, this field takes priority over question. User question to query against the knowledge base. Optional field. Set to 'QuestionOnly' for using a question only Ranker. Threshold for answers returned based on score. Find only answers that contain these metadata. # Set the `runtime_endpoint` variable to your QnA Maker runtime endpoint. # The value of `runtime_endpoint` has the format . * Create a knowledgebase * Update a knowledgebase * Publish a knowledgebase, waiting for publishing to complete * Get Query runtime endpoint key * Download a knowledgebase * Get answer * Delete a knowledgebase This article demonstrates how to call a REST API endpoint for QnA Maker, a service in Azure Cognitive Services suite. QnA Maker is a cloud-based Natural Language Processing (NLP) service that allows you to create a natural conversational layer over your data. Use of the QnA program is straightforward. Simply type Q: followed by a relevance clause, and click the Q/A button for evaluation. The QnA program can evaluate many queries at the same time. It ignores any text not preceded by Q:. To query by using QnA, click Evaluate > Evaluate Using > Query Channel using QnA. To query in client context, click Evaluate > Evaluate Using > Query Channel using Client. In the login window opened by the Fixlet Debugger, enter the credentials to connect to a BigFix environment. The input file would be a series of relevance queries, one per line, including the Q: at the front. I’m using this method from a Scheduled Task to run a query and take some actions based on the query output. The precise answering feature introduced in QnA Maker managed (Preview), allows you to get the precise short answer from the best candidate answer passage present in the knowledge-base for any user query. When utilizing the azurerm_cognitive_account and specifying the kind as QnAMaker the qna_runtime_endpoint flag is a required. I have tried to utilize this but I have run into issues and not been able to find any documentation on what information should be added to the qna_runtime_endpoint.