{"data":[{"active":true,"blog_title":null,"blog_url":null,"feature_description":"Fabric feature store is a centralized system that manages the data inputs, called features, that power machine learning models. It bridges raw data and ML models, giving teams one place to define, store, serve, and discover features instead of each team building its own pipelines.","feature_name":"Feature Store","last_modified":"2026-08-11","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q4 2026","release_item_id":"75004e12-2158-f111-bec7-6045bd00f798","release_status":"Planned","release_type":"Public preview"},{"active":true,"blog_title":null,"blog_url":null,"feature_description":"ML Endpoints are currently available for a limited set of ML model flavors, including Keras, LightGBM, Sklearn, and XGBoost. This feature allows users to activate ML endpoints for models with tensor-based schemas or no schemas.","feature_name":"Support tensor based models on model endpoints","last_modified":"2026-07-14","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q3 2026","release_item_id":"57c0f1aa-2358-f111-bec7-6045bd00f798","release_status":"Planned","release_type":"Public preview"},{"active":true,"blog_title":"Unlock insights from images and PDFs with multimodal support in Fabric AI functions (Preview)","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/unlock-insights-from-images-and-pdfs-with-multimodal-support-in-fabric-ai-functi/5172205","feature_description":"Multimodal support for AI functions in Fabric will allow notebook users to use AI functions on their unstructured data in pdf, image, and text file formats, to seamlessly perform tasks like summarization, classification, sentiment analysis, and more, with just a few lines of code. The multimodal support for AI functions will be available with both pandas and Spark.","feature_name":"Multimodal Support for AI Functions [Public Preview]","last_modified":"2026-07-02","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q1 2026","release_item_id":"cda45361-58b3-f011-bbd3-000d3a3740cc","release_status":"Shipped","release_type":"Public preview"},{"active":true,"blog_title":"Announcing low code Automated ML (AutoML) in Fabric Data Science","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/announcing-low-code-automated-ml-automl-in-fabric-data-science/5172992","feature_description":"Automated Machine Learning (AutoML) enables users to build and deploy machine learning models by automating the most time-consuming and complex parts of the model development process. The General Availability will enable users to simplofy the ML development cycle by using AutoML to automatically select the best algorithms, tune hyperparameters, and generate optimized models based on the input data and desired outcomes.","feature_name":"GA AutoML in Fabric","last_modified":"2026-07-02","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q1 2026","release_item_id":"2b64b096-5eb3-f011-bbd3-000d3a30273e","release_status":"Shipped","release_type":"General availability"},{"active":true,"blog_title":"Fabric AI Functions Enhancements (Generally Available)","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/fabric-ai-functions-enhancements-generally-available/5172479","feature_description":"AI functions in Fabric will allow notebook users to seamlessly perform tasks like text summarization, translation, classification, sentiment analysis, grammar correction, and more, providing a simplified API for common enrichments and making it easier for users to apply them with fewer lines of code. The functions will be initially available on top of pandas DataFrames and ultimately available via Spark, SQL, and other programming surfaces across Fabric.","feature_name":"AI Functions for LLM-Powered Text Enrichment and Transformation [GA]","last_modified":"2026-07-02","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q4 2025","release_item_id":"9f3e8191-53b3-f011-bbd3-000d3a3740cc","release_status":"Shipped","release_type":"General availability"},{"active":true,"blog_title":"Accelerate Data Transformation with AI Functions in Data Wrangler (Preview)","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/accelerate-data-transformation-with-ai-functions-in-data-wrangler-preview/5172559","feature_description":"AI functions in Data Wrangler will allow users to enrich and transform data directly within the Data Wrangler interface. With just a few clicks, users can perform tasks such as text summarization, translation, classification, sentiment analysis, grammar correction, and more, streamlining data preparation through seamless, built-in AI capabilities.","feature_name":"AI functions in Data Wrangler [Public Preview]","last_modified":"2026-07-02","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q3 2025","release_item_id":"c638bc4d-6f5e-f011-bec2-0022480939f0","release_status":"Shipped","release_type":"Public preview"},{"active":true,"blog_title":"Accelerate Data Transformation with AI Functions in Data Wrangler (Preview)","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/accelerate-data-transformation-with-ai-functions-in-data-wrangler-preview/5172559","feature_description":"A new suite of AI-powered operations in Data Wrangler will allow users to describe code transformations with natural language and generate the corresponding Python; translate custom Python code into PySpark code; and view high-confidence suggestions for operations based on their working data.","feature_name":"Low-Code AI-Powered Operations in Data Wrangler [GA]","last_modified":"2026-07-02","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q3 2025","release_item_id":"053c13db-53b3-f011-bbd3-000d3a3740cc","release_status":"Shipped","release_type":"General availability"},{"active":true,"blog_title":"Announcing AI functions for seamless data engineering with GenAI","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/announcing-ai-functions-for-seamless-data-engineering-with-genai/5172902","feature_description":"AI functions in Fabric will allow notebook users to seamlessly perform tasks like text summarization, translation, classification, sentiment analysis, grammar correction, and more, providing a simplified API for common enrichments and making it easier for users to apply them with fewer lines of code. The functions will be initially available on top of pandas DataFrames and ultimately available via Spark, SQL, and other programming surfaces across Fabric.","feature_name":"AI Functions for LLM-Powered Text Enrichment and Transformation [Public Preview]","last_modified":"2026-07-02","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q1 2025","release_item_id":"39d793d1-b79c-ef11-8a6a-002248098a98","release_status":"Shipped","release_type":"Public preview"},{"active":true,"blog_title":"Announcing low code Automated ML (AutoML) in Fabric Data Science","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/announcing-low-code-automated-ml-automl-in-fabric-data-science/5172992","feature_description":"Our Low-Code AutoML tool empowers data scientists and analysts to easily create machine learning models without the need for extensive coding. Through an intuitive, step-by-step wizard, users can configure and launch AutoML trials directly from the user interface.","feature_name":"Low Code AutoML","last_modified":"2026-07-02","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q4 2024","release_item_id":"a05607d1-c758-ef11-a317-0022480abf3c","release_status":"Shipped","release_type":"Public preview"},{"active":true,"blog_title":null,"blog_url":null,"feature_description":"This feature is to offer a model alias that provides a stable, human-readable name pointing to a specific model version. This lets them switch, promote, or roll back model versions safely without breaking downstream integrations.","feature_name":"MLOps: model alias","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q3 2026","release_item_id":"761be85d-2358-f111-bec7-6045bd00f798","release_status":"Planned","release_type":"Public preview"},{"active":true,"blog_title":"Empowering agentic AI by integrating Fabric with Azure AI Foundry","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/empowering-agentic-ai-by-integrating-fabric-with-azure-ai-foundry/5172877","feature_description":"Formerly named Azure Cognitive Services, Foundry Tools on Fabric brings in Azure Language, Azure Translator, and Azure Content Understanding as native endpoints for both Fabric workload teams and end users. With these endpoints, Fabric workload teams and end users can build intelligent solutions without having AI or data science expertise.","feature_name":"GA foundry tools in Fabric","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q2 2026","release_item_id":"be5a72e9-3c53-f111-bec7-6045bd00fc61","release_status":"Planned","release_type":"General availability"},{"active":true,"blog_title":"Cross-workspace logging for MLflow in Microsoft Fabric: Build MLOps workflows with confidence (Generally Available)","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/cross-workspace-logging-for-mlflow-in-microsoft-fabric-build-mlops-workflows-wit/5172000","feature_description":"MLflow 3 arrives in Fabric with support for tracing, logged models, and an upgraded experiment tracking UX. The update introduces a new trace table view and nested trace visualization for inspecting multi-step model execution paths, a dedicated trace detail page, and the logged model API and UX for capturing and managing model artifacts as first-class objects. Together, these provide a more structured workflow for tracking experiments, registering models, and understanding model behavior.","feature_name":"MLFlow 3 in Fabric : logged model and traces","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q2 2026","release_item_id":"39e50387-3c53-f111-bec7-6045bd00fc61","release_status":"Planned","release_type":"Public preview"},{"active":true,"blog_title":"Outbound access protection for Data Factory (Generally Available)","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/outbound-access-protection-for-data-factory-generally-available/5172006","feature_description":"This feature enables Fabric enterprise customers to use ML models in Outbound Access Protection (OAP) enabled workspace and prevent sensitive data from getting exfiltrated.","feature_name":"Outbound Access Protection for ML Model","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q1 2026","release_item_id":"e3b77c88-c4bb-f011-bbd3-6045bd00f9db","release_status":"Shipped","release_type":"General availability"},{"active":true,"blog_title":"Outbound access protection for Data Factory (Generally Available)","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/outbound-access-protection-for-data-factory-generally-available/5172006","feature_description":"This feature allow fabric enterprise customers to use ML experiments in Outbound Access Protection enabled workspace and prevent sensitive data from getting exfiltrated.","feature_name":"Outbound Access Protection for ML Experiment","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q1 2026","release_item_id":"b5548630-c4bb-f011-bbd3-6045bd00f9db","release_status":"Shipped","release_type":"General availability"},{"active":true,"blog_title":"Serve real-time predictions seamlessly with ML model endpoints","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/serve-real-time-predictions-seamlessly-with-ml-model-endpoints/5172668","feature_description":"After activating ML Model endpoints in Fabric, users will need to monitor the quality of their models and the health of their endpoints in the artifact UI with a corresponding set of dynamic graphs. With this feature released, users are able to see request rate, request error rate, and latency on ML model endpoint monitoring UI.","feature_name":"Machine Learning Model Endpoint Monitoring","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q1 2026","release_item_id":"af0622f7-13be-f011-bbd3-000d3a30273e","release_status":"Shipped","release_type":"Public preview"},{"active":true,"blog_title":"Enhancing Open Source: Fabric\u2019s Contributions to FLAML for Scalable AutoML","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/enhancing-open-source-fabrics-contributions-to-flaml-for-scalable-automl/5173064","feature_description":"This feature enables users to deploy models with internal dependencies, such as AutoML or FLAML, directly through ML model endpoints.","feature_name":"Support AutoML with FLAML in ML model endpoints","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q4 2025","release_item_id":"f282e269-15be-f011-bbd3-000d3a30273e","release_status":"Shipped","release_type":"Public preview"},{"active":true,"blog_title":"Microsoft Fabric Lifecycle Management \u2013 Getting started with Git Integration and Deployment Pipelines","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/microsoft-fabric-lifecycle-management---getting-started-with-git-integration-and/5173247","feature_description":"The Machine learning experiments and models integrate with the lifecycle management capabilities in Microsoft Fabric, providing a standardized collaboration between all development team members throughout the product's life.  https://learn.microsoft.com/en-us/fabric/data-science/machine-learning-artifacts-git-deployment-pipelines","feature_name":"CI/CD for Models and Experiments","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q3 2025","release_item_id":"bc4db3eb-ed23-f011-8c4d-000d3a34671f","release_status":"Shipped","release_type":"Public preview"},{"active":true,"blog_title":"Serve real-time predictions seamlessly with ML model endpoints","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/serve-real-time-predictions-seamlessly-with-ml-model-endpoints/5172668","feature_description":"Alongside the existing functionality for batch scoring with PREDICT, Fabric will allow data scientists to serve real-time predictions from any registered ML Model using secure, scalable online endpoints that are automatically configured. These endpoints can be called from other Fabric engines or from external apps, allowing users to deploy their models for wide, reliable consumption.","feature_name":"Real-Time Endpoints for Machine Learning Models [Public Preview]","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q3 2025","release_item_id":"a3b61e19-5da0-ef11-8a6a-00224804e9b4","release_status":"Shipped","release_type":"Public preview"},{"active":true,"blog_title":"Service Principal Support in Semantic Link: Enabling Scalable, Secure Automation","blog_url":"https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/service-principal-support-in-semantic-link-enabling-scalable-secure-automation/5172549","feature_description":"This feature adds Azure service principal (SPN) support as an authentication type for Semantic Link.An Azure service principal is a non-human security identity used by applications or automation tools to access Azure resources with precise permissions.Using service principals in Semantic Link allows Power BI admins to manage and optimize Fabric items securely and efficiently, without relying on user identities.","feature_name":"Service Principal authentication supported in Semantic Link","last_modified":"2026-06-16","product_id":"0522b590-dcd6-ee11-9079-000d3a310f67","product_name":"Data Science","release_date":"Q2 2025","release_item_id":"e399e3d4-b624-f011-8c4d-00224804b6c3","release_status":"Shipped","release_type":"Public preview"}],"links":{"first":"/api/releases?product_name=Data+Science&page_size=50&page=1","last":"/api/releases?product_name=Data+Science&page_size=50&page=1","next":null,"prev":null,"self":"/api/releases?product_name=Data+Science&page_size=50&page=1"},"pagination":{"has_next":false,"has_prev":false,"next_page":null,"page":1,"page_size":50,"prev_page":null,"total_items":19,"total_pages":1}}