Merge pull request #272244 from aahill/model-update

assistants faq

Author: v-dirichards

articles/ai-services/openai/concepts/assistants.md

@@ -25,6 +25,9 @@ Assistants API supports persistent automatically managed threads. This means tha
 - [Code Interpreter](../how-to/code-interpreter.md)
 - [Function calling](../how-to/assistant-functions.md)
 
+> [!TIP]
+> There is no additional [pricing](https://azure.microsoft.com/pricing/details/cognitive-services/openai-service/) or [quota](../quotas-limits.md) for using Assistants unless you use the [code interpreter](../how-to/code-interpreter.md) tool.
+
 Assistant API is built on the same capabilities that power OpenAI’s GPT product. Some possible use cases range from AI-powered product recommender, sales analyst app, coding assistant, employee Q&A chatbot, and more. Start building on the no-code Assistants playground on the Azure OpenAI Studio or start building with the API.
 
 > [!IMPORTANT]

articles/ai-services/openai/faq.yml

@@ -118,7 +118,7 @@ sections:
         answer:
           If the service performs processing, you will be charged even if the status code is not successful (not 200).
           Common examples of this are, a 400 error due to a content filter or input limit, or a 408 error due to a timeout. Charges will also occur when a `status 200` is received with a `finish_reason` of `content_filter`.
-          In this case the prompt did not have any issues, but the completion generated by the model was detected to violate the content filtering rules which results in the completion being filtered. 
+          In this case the prompt did not have any issues, but the completion generated by the model was detected to violate the content filtering rules, which result in the completion being filtered. 
 
           If the service doesn't perform processing, you won't be charged.
           For example, a 401 error due to authentication or a 429 error due to exceeding the Rate Limit.
@@ -228,7 +228,58 @@ sections:
           What are the known limitations of GPT-4 Turbo with Vision?
         answer: |
           See the [limitations](./concepts/gpt-with-vision.md#limitations) section of the GPT-4 Turbo with Vision concepts guide.
+  - name: Assistants
+    questions:
+      - question: |
+          Do you store any data used in the Assistants API? 
+        answer: |
+          Yes. Unlike Chat Completions API, Azure OpenAI Assistants is a stateful API, meaning it retains data. There are two types of data stored in the Assistants API: 
+          * Stateful entities: Threads, messages, and runs created during Assistants use. 
+          * Files: Uploaded during Assistants setup or as part of a message.  
+      - question: |
+          Where is this data stored?
+        answer: |
+          Data is stored in a secure, Microsoft-managed storage account that is logically separated.  
+      - question: |
+          How long is this data stored?
+        answer: |
+          All used data persists in this system unless you explicitly delete this data. Use the [delete function](./assistants-reference-threads.md) with the thread ID of the thread you want to delete. Clearing the Run in the Assistants Playground does not delete threads, however deleting them using delete function will not list them in the thread page. 
+      - question: |
+          Can I bring my own data store to use with Assistants? 
+        answer: |
+          No. Currently Assistants supports only local files uploaded to the Assistants-managed storage. You cannot use your private storage account with Assistants. 
+      - question: |
+          Is my data used by Microsoft for training models? 
+        answer: |
+          No. Data is not used for Microsoft not used for training models. See the [Responsible AI documentation](/legal/cognitive-services/openai/data-privacy?context=%2Fazure%2Fai-services%2Fopenai%2Fcontext%2Fcontext) for more information. 
+      - question: |
+          Where is data stored geographically?
+        answer: |
+          Azure OpenAI Assistants endpoints are regional, and data is stored in the same region as the endpoint. For more information, see the [Azure data residency documentation](https://azure.microsoft.com/explore/global-infrastructure/data-residency/#overview). 
+      - question: |
+          How am I charged for Assistants?
+        answer: |
+          Currently, when you use Assistants API, you're billed for the following: 
+          - Inference cost (input and output) of the base model you're using for each Assistant (for example gpt-4-0125). If you've created multiple Assistants, you will be charged for the base model attached to each Assistant. 
+          - If you've enabled the Code Interpreter tool. For example if your assistant calls Code Interpreter simultaneously in two different threads, this would create two Code Interpreter sessions, each of which would be charged. Each session is active by default for one hour, which means that you would only pay this fee once if your user keeps giving instructions to Code Interpreter in the same thread for up to one hour. 
 
+          For more information, see the [pricing page](https://azure.microsoft.com/pricing/details/cognitive-services/openai-service/).
+      - question: |
+          Is there any additional pricing or quota for using Assistants? 
+        answer: |
+          No. All [quotas](./quotas-limits.md) apply to using models with Assistants.
+      - question: |
+          Does the Assistants API support non-Azure OpenAI models?   
+        answer: |
+          Assistants API only supports Azure OpenAI models.
+      - question: |
+          Is the Assistants API generally available?
+        answer: |
+          The Assistants API is currently in public preview. Stay informed of our latest product updates by regularly visiting our [What's New](./whats-new.md) page.
+      - question: |
+          What are some examples or other resources I can use to learn about Assistants?  
+        answer: |
+          See the [Conceptual](./concepts/assistants.md), [quickstart](./assistants-quickstart.md), [how-to](./how-to/assistant.md) articles for information on getting started and using Assistants. You can also check out Azure OpenAI Assistants code samples on [GitHub](https://github.com/Azure-Samples/azureai-samples/tree/main/scenarios/Assistants).
   - name: Web app
     questions:
       - question: |
@@ -260,7 +311,7 @@ sections:
       - question: |
           How can I customize or automate the index creation process?
         answer:
-          You can prepare the index yourself using a [script provided on GitHub](https://go.microsoft.com/fwlink/?linkid=2244395). Using this script will create an Azure AI Search index with all the information needed to better leverage your data, with your documents broken down into manageable chunks. Please see the README file with the data preparation code for details on how to run it.
+          You can prepare the index yourself using a [script provided on GitHub](https://go.microsoft.com/fwlink/?linkid=2244395). Using this script will create an Azure AI Search index with all the information needed to better use your data, with your documents broken down into manageable chunks. See the README file with the data preparation code for details on how to run it.
       - question: | 
           How can I update my index?
         answer:
@@ -289,7 +340,7 @@ sections:
           If Semantic Search is enabled for my Azure AI Search resource, will it be automatically applied to Azure OpenAI on your data in the Azure OpenAI Studio?
         answer:
           When you select "Azure AI Search" as the data source, you can choose to apply semantic search. 
-          If you select "Azure Blob Container" or "Upload files" as the data source, you can create the index as usual. Afterwards you would re-ingest the data using the "Azure AI Search" option to select the same index and apply Semantic Search. You will then be ready to chat on your data with semantic search applied.
+          If you select "Azure Blob Container" or "Upload files" as the data source, you can create the index as usual. Afterwards you would reingest the data using the "Azure AI Search" option to select the same index and apply Semantic Search. You will then be ready to chat on your data with semantic search applied.
       - question: |
           How can I add vector embeddings when indexing my data?
         answer: