Merge branch 'patch-14' into kate-4954-4949-overview-overview

Author: laujan

articles/ai-foundry/concepts/models-featured.md

@@ -250,7 +250,7 @@ See [the Microsoft model collection in Azure AI Foundry portal](https://ai.azure
 
 Mistral AI offers two categories of models, namely: 
 
-- _Premium models_: These include Mistral Large, Mistral Small, Mistral-OCR-2503, and Ministral 3B models, and are available as serverless APIs with pay-as-you-go token-based billing.  
+- _Premium models_: These include Mistral Large, Mistral Small, Mistral-OCR-2503, Mistral Medium 3 (25.05), and Ministral 3B models, and are available as serverless APIs with pay-as-you-go token-based billing.  
 - _Open models_: These include Mistral-small-2503, Codestral, and Mistral Nemo (that are available as serverless APIs with pay-as-you-go token-based billing), and [Mixtral-8x7B-Instruct-v01, Mixtral-8x7B-v01, Mistral-7B-Instruct-v01, and Mistral-7B-v01](../how-to/deploy-models-mistral-open.md)(that are available to download and run on self-hosted managed endpoints).
 
 
@@ -262,6 +262,7 @@ Mistral AI offers two categories of models, namely:
 | [Mistral-Large-2411](https://ai.azure.com/explore/models/Mistral-Large-2411/version/2/registry/azureml-mistral) | [chat-completion](../model-inference/how-to/use-chat-completions.md?context=/azure/ai-foundry/context/context) |  - **Input:** text (128,000 tokens) <br /> - **Output:** text (4,096 tokens) <br /> - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
 | [Mistral-large-2407](https://ai.azure.com/explore/models/Mistral-large-2407/version/1/registry/azureml-mistral) <br /> (deprecated) | [chat-completion](../model-inference/how-to/use-chat-completions.md?context=/azure/ai-foundry/context/context) |  - **Input:** text (131,072 tokens) <br /> - **Output:** text (4,096 tokens) <br /> - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON  |
 | [Mistral-large](https://ai.azure.com/explore/models/Mistral-large/version/1/registry/azureml-mistral) <br /> (deprecated) | [chat-completion](../model-inference/how-to/use-chat-completions.md?context=/azure/ai-foundry/context/context) |  - **Input:** text (32,768 tokens) <br /> - **Output:** text (4,096 tokens) <br />  - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
+| [Mistral-medium-2505](https://aka.ms/aistudio/landing/mistral-medium-2505) | [chat-completion](../model-inference/how-to/use-chat-completions.md?context=/azure/ai-foundry/context/context) |  - **Input:** text (128,000 tokens), image <br /> - **Output:** text (128,000 tokens) <br />  - **Tool calling:** No <br /> - **Response formats:** Text, JSON |
 | [Mistral-OCR-2503](https://aka.ms/aistudio/landing/mistral-ocr-2503) | [image to text](../how-to/use-image-models.md) |  - **Input:** image or PDF pages (1,000 pages, max 50MB PDF file) <br> - **Output:** text <br /> - **Tool calling:** No <br /> - **Response formats:** Text, JSON, Markdown |
 | [Mistral-small-2503](https://aka.ms/aistudio/landing/mistral-small-2503) | [chat-completion (with images)](../model-inference/how-to/use-chat-multi-modal.md?context=/azure/ai-foundry/context/context) |  - **Input:** text and images (131,072 tokens), <br> image-based tokens are 16px x 16px <br> blocks of the original images <br /> - **Output:** text (4,096 tokens) <br /> - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |
 | [Mistral-small](https://ai.azure.com/explore/models/Mistral-small/version/1/registry/azureml-mistral) | [chat-completion](../model-inference/how-to/use-chat-completions.md?context=/azure/ai-foundry/context/context) |  - **Input:** text (32,768 tokens) <br /> - **Output:** text (4,096 tokens) <br /> - **Tool calling:** Yes <br /> - **Response formats:** Text, JSON |

articles/ai-foundry/includes/region-availability-maas.md

@@ -96,6 +96,7 @@ Codestral-2501    | [Microsoft Managed Countries/Regions](/partner-center/market
 Mistral OCR 25.03   | [Microsoft Managed Countries/Regions](/partner-center/marketplace/tax-details-marketplace#microsoft-managed-countriesregions)   <br> Brazil <br> Hong Kong SAR <br> Israel      | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3           |  Not available       |
 Mistral Nemo  | [Microsoft Managed Countries/Regions](/partner-center/marketplace/tax-details-marketplace#microsoft-managed-countriesregions)  <br> Brazil <br> Hong Kong SAR <br> Israel     | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3          | East US 2 <br> East US <br> North Central US <br> South Central US <br> West US <br> West US 3    |
 Mistral Small 25.03 <br> Mistral Small    | [Microsoft Managed Countries/Regions](/partner-center/marketplace/tax-details-marketplace#microsoft-managed-countriesregions)   <br> Brazil <br> Hong Kong SAR <br> Israel      | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3           |  Not available       |
+Mistral Medium 3 (25.05)    | [Microsoft Managed Countries/Regions](/partner-center/marketplace/tax-details-marketplace#microsoft-managed-countriesregions)   <br> Brazil <br> Hong Kong SAR <br> Israel      | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3           |  Not available       |
 Ministral-3B | [Microsoft Managed Countries/Regions](/partner-center/marketplace/tax-details-marketplace#microsoft-managed-countriesregions)  <br> Brazil <br> Hong Kong SAR<br> Israel    | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3           | East US 2 <br> East US <br> North Central US <br> South Central US <br> West US <br> West US 3       |
 Mistral Large (2407) <br> Mistral-Large    | [Microsoft Managed Countries/Regions](/partner-center/marketplace/tax-details-marketplace#microsoft-managed-countriesregions)  <br> Brazil <br> Hong Kong SAR<br> Israel    | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3           | Not available       |
 Mistral-Large (2411)  | [Microsoft Managed Countries/Regions](/partner-center/marketplace/tax-details-marketplace#microsoft-managed-countriesregions)  <br> Brazil <br> Hong Kong SAR<br> Israel    | East US <br> East US 2 <br> North Central US <br> South Central US <br> Sweden Central <br> West US <br> West US 3           | East US 2 <br> East US <br> North Central US <br> South Central US <br> West US <br> West US 3       |

articles/ai-services/content-understanding/concepts/standard-pro-modes.md

@@ -82,10 +82,14 @@ Not sure which mode is right for your scenario? The following charts compare sta
 
 You can try out the features of both Content Understanding standard and pro modes using the [Azure AI Foundry](https://ai.azure.com/explore/aiservices/vision/contentunderstanding). The service enables you to bring your own data and experiment with all the functionalities of both modes in a lightweight, no-code approach to help you find the best fit for your unique scenario.
 
-### Pro mode known limitations
+### Pro mode known limitations and best practices
 
 * Content Understanding pro mode currently doesn't offer confidence scores or grounding. It currently supports generative and classification of your fields but doesn't support extraction only.
 * Content Understanding pro mode is currently only available for documents.
+* The system works in “lookup mode” on reference documents: don’t expect exhaustive information recovery; if this is desired, include the document in your input documents instead.
+* Be as specific as possible with your schema (e.g., rather than identifying a list of inconsistencies, create a list field for each inconsistency type and describe it), and if possible, reference which parts of which documents should be consulted.
+* Keep reference documents short and sweet: Limit documents to the critical ones, and keep them short where possible to improve recall.
+
 
 ## Next steps
 

articles/ai-services/content-understanding/document/markdown.md

@@ -17,7 +17,7 @@ Azure AI Content Understanding converts unstructured documents into [GitHub Flav
 
 ## Words and selection marks
 
-Recognized words and detected selection marks are represented in markdown as plain text. Content may be escaped to avoid ambiguity with markdown formatting syntax.
+Recognized words and detected selection marks are represented in markdown as plain text. Selection marks are encoded using the Unicode characters `☒` (selected) and `☐` (unselected). Content might be escaped to avoid ambiguity with markdown formatting syntax.
 
 ## Barcodes
 
@@ -106,7 +106,7 @@ This is a footnote.
 ## Page metadata
 
 Markdown doesn't natively encode page metadata, such as page numbers, headers, footers, and breaks.
-Since this information may be useful for downstream applications, we encode such metadata as HTML comments.
+Since this information might be useful for downstream applications, we encode such metadata as HTML comments.
 
 | Metadata | Markdown |
 | --- | --- |

articles/ai-services/content-understanding/document/overview.md

@@ -98,4 +98,4 @@ Developers using Content Understanding should review Microsoft's policies on cus
 * Try processing your document content using Content Understanding in [Azure AI Foundry](https://aka.ms/cu-landing).
 * Learn to analyze document content [**analyzer templates**](../quickstart/use-ai-foundry.md).
 * Review code samples: [**visual document search**](https://github.com/Azure-Samples/azure-ai-search-with-content-understanding-python/blob/main/notebooks/search_with_visual_document.ipynb).
-* Review code sample: [**analyzer templates**](https://github.com/Azure-Samples/azure-ai-content-understanding-python/tree/main/analyzer_templates).
+

articles/ai-services/content-understanding/glossary.md

@@ -20,11 +20,11 @@ ms.author: lajanuar
 | **Analyzer template** | A predefined configuration and field schema for an analyzer. It simplifies creating analyzers by allowing modifications to a template instead of starting from scratch. This feature is available only in [Azure AI Foundry portal](https://ai.azure.com/), not via REST API/SDKs. |
 | **Analyzer result** | The output generated by an analyzer after processing input data. It typically includes extracted content in Markdown, extracted fields, and optional modality-specific details. |
 | **Add-ons** | Added features that enhance content extraction results, such as layout elements, barcodes, and figures in documents. |
-| **Fields** | List of structured key-value pairs derived from the content, as defined by the field schema. [Learn more about supported field value types.](service-limits.md) |
+| **Fields** | List of structured key-value pairs derived from the content, as defined by the field schema. [Learn more about supported field value types.](service-limits.md#field-schema-limits) |
 | **Field schema** | A formal description of the fields to extract from the input. It specifies the name, description, value type, generation method, and more for each field. |
 | **Generation method** | The process of determining the extracted value of a specified field. Content Understanding supports: <br/> &bullet; **Extract**: Directly extract values from the input content, such as dates from receipts or item details from invoices. <br/> &bullet; **Classify**: Classify content into predefined categories, such as call sentiment or chart type. <br/> &bullet; **Generate**: Generate values from input data, such as summarizing an audio conversation or generating scene descriptions from videos. |
 | **Span** | A reference indicating the location of an element (for example, field, word) within the extracted Markdown content. A character offset and length represent a span. Different programming languages use various character encodings, which can affect the exact offset and length values for Unicode text. To avoid confusion, spans are only returned if the desired encoding is explicitly specified in the request. Some elements can map to multiple spans if they aren't contiguous in the markdown (for example, page). |
-| **Processing Location** | An API request parameter that defines the geographic region where Azure AI Services analyzes your data. You can choose from three options: `geography`, `dataZone`, and `global` to control where processing occurs. This setting helps meet data residency requirements and optimize performance or scalability based on your needs. For more information, *see* the Content Understanding API reference documentation.
+| **Processing Location** | An API request parameter that defines the geographic region where Azure AI Services analyzes your data. You can choose from three options: `geography`, `dataZone`, and `global` to control where processing occurs. This setting helps meet data residency requirements and optimize performance or scalability based on your needs.
 | **Grounding source** | The specific regions in content where a value was generated. It has different representations depending on the file type: <br>&bullet; **Image** - A polygon in the image, often an axis-aligned rectangle (bounding box). <br>&bullet; **PDF/TIFF** - A polygon on a specific page, often a quadrilateral. <br>&bullet; **Audio** - A start and end time range. <br>&bullet; **Video** - A start and end time range with an optional polygon in each frame, often a bounding box.|
 | **Person directory** | A structured way to store face data for recognition tasks. You can add individual faces to the directory and later search for visually similar faces. You can also create person profiles, associate faces to them, and match new face images to known individuals. This setup supports both flexible face matching and identity recognition across images and videos. |
 | **Confidence score** | The level of certainty that the extracted data is accurate. |

articles/ai-services/content-understanding/image/overview.md

@@ -42,12 +42,10 @@ Content Understanding offers several key benefits for extracting information fro
 
 ## Get started
 
-Get started with processing images with Content Understanding by following our [REST API quickstart](LINK TO IMAGE TAB) or visiting [Azure AI Foundry](https://aka.ms/cu-landing) for a no code experience. 
+Get started with processing images with Content Understanding by following our [REST API quickstart](../quickstart/use-rest-api.md?tabs=image) or visiting [Azure AI Foundry](https://aka.ms/cu-landing) for a no code experience. 
 
 > [!NOTE]
-> For best results, image schema should only be used to process non-document-based images.
-> Text heavy images of documents should be processed using a document schema.
-> Use cases that require extraction of text from document images or scanned documents should be processed using a document field extraction schema.
+> Image analyzers are currently not optimized for scenarios where analysis is based primarily on extracted text. If your main goal is to extract and analyze text from images, consider using a document field extraction schema instead.
 
 
 > [!IMPORTANT]

articles/ai-services/content-understanding/whats-new.md

@@ -14,22 +14,23 @@ ms.custom:
 
 # What's new in Azure AI Content Understanding?
 
-The Azure AI Content Understanding service is continuously updated. Bookmark this page to stay informed about the latest features and samples.
+Azure AI Content Understanding service is updated on an ongoing basis. Bookmark this page to stay up to date with release notes, feature enhancements, and our newest documentation.
 
 ## May 2025
 The Azure AI Content Understanding **`2025-05-01-preview`** REST API is now available. This update introduces the following updates and enhanced capabilities:
 
-* **Modes for documents**: With the **`2025-05-01-preview`** release, we introduce two modes: `standard` and `pro`. The `pro` mode, currently exclusive to the document analyzer, enables advanced capabilities. Content Understanding now supports reasoning across multiple documents as input for external knowledge, empowering users to derive agentic inferences directly from reference documents.
+* **Modes for documents**: With the **`2025-05-01-preview`** release, we introduce two modes: `standard` and `pro`. Content Understanding pro mode adds reasoning, support for multiple input documents, the ability to configure an external knowledge base for linking and validation. These features enable agentic processes for automating complex tasks requiring human effort. The `pro` mode is currently limited to the document analyzer, enables reasoning capabilities that can infer complicated outputs given multiple reference documents (for example, insurance policy documents) and multiple input documents (for example, forms and supporting documents that are case-specific)
 
 * **Document modality improvements** :
 
-  * Support for cross table extraction spanning multiple pages.
-  * Selection mark support for checkmark and radio buttons as unicode characters.
-  * Confidence score improvements.
-  * Support for embedded figures in Markdown as base64 string format.
+  * Improvements in content extraction:
+      * Support for table spanning multiple pages getting extracted as a single logical table. 
+      * Selection mark support for checkmark and radio buttons as unicode characters.
+      * Bar code extraction as part of the default content extraction along with `OCR`.
+      * Support for embedded figures in Markdown as base64 string format output.
+  * Confidence score improvements with better grounding results for extractive fields. 
   * New file format support extended for following document types `.rtf`,`.txt`,`.xml`,`.json`, `.msg`,`.eml`,`.csv`, and `.xlsx`.
-  * Bar code extraction as part of the default content extraction along with `OCR`.
-  * Classification API for documents only. This API supports classifying and splitting documents with optional routing to analyzers.
+  * Classification API for documents only. This API supports classifying and splitting a single file containing multiple documents  with optional routing to field extraction analyzers.
 
 * **Video Modality improvements**:
 
@@ -43,10 +44,12 @@ The Azure AI Content Understanding **`2025-05-01-preview`** REST API is now avai
 * **Face API preview**: Detection, recognition, and enrollment of faces.
 
 * **Billing Changes**: New simpler pricing model that lowers costs from the **2024-12-01-preview** REST API across all modalities.
+  
+* **User experience improvements**: To improve latency issues, the UX experience is refreshed with ease-of-use and quick try-out experience as part of [Azure AI Foundry](https://aka.ms/cu-landing)
 
 ## April 2025
 
-The Azure AI Content Understanding **2024-12-01-preview** REST API is now available. This update for preview API introduces the following updates and enhanced capabilities:
+**2024-12-01-preview** REST API introduces the following updates and enhanced capabilities:
 
 * **General improvements**. For all modality, to request an increase from current limits, contact us at `[email protected]`.
 * **Prebuilt invoice template**. The invoice template is now customizable. Once you select the invoice template, you can access a predefined list of fields that can be tailored to your specific needs by adding or removing fields.