Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
This article is continuously updated with a recent review of what's new in Microsoft Fabric.
- To stay up to date with the latest news and features, see the Microsoft Fabric Updates Blog.
- For community, marketing, case studies, and industry news, see the Microsoft Fabric Blog.
- For the latest in Power BI, see What's new in Power BI?
- For older updates, see the Microsoft Fabric What's New archive.
New to Microsoft Fabric?
- Learning Paths for Fabric
- Get started with Microsoft Fabric
- End-to-end tutorials in Microsoft Fabric
- Microsoft Fabric terminology
Features currently in preview
The following table lists the features of Microsoft Fabric that are currently in preview. Preview features are sorted alphabetically.
Note
Features currently in preview are available under supplemental terms of use. Review the legal terms that apply to Azure features that are in beta, preview, or otherwise not yet released into general availability. Microsoft Fabric provides previews to give you a chance to evaluate and share feedback with the product group on preview features before they become generally available (GA).
| Feature | Learn more |
|---|---|
| AI functions in Fabric Data Warehouse (preview) | Fabric Data Warehouse now brings AI directly into T-SQL, enabling you to categorize and classify text, analyze sentiment, extract structured information, translate text across languages, and even correct grammar—all using built-in AI functions. To get started, see Use AI functions in Fabric Data Warehouse (preview). |
| Multimodal support in Fabric AI functions (Preview) | AI functions can process images (JPG/JPEG, PNG, GIF, WebP), PDFs, and common text formats alongside text data. Most functions accept file-path inputs via column_type="path". New helpers include aifunc.load (folder-to-table ingestion with optional prompt and schema), aifunc.list_file_paths (file URL enumeration), and ai.infer_schema (schema inference compatible with ai.extract). For more information, see Transform and enrich data with AI functions and Extract information (ai.extract). |
| Access your Delta Lake tables as Apache Iceberg in OneLake (Preview) | OneLake now lets you access your Delta Lake tables using Apache Iceberg compatible readers, automatically, without data movement or duplication. To get started, see Use Iceberg tables with OneLake. |
| Activator as business events publisher (Preview) | Activator provides a no-code way to publish business events in Microsoft Fabric. When Activator detects that a condition is met in your data, it can emit a structured business event into Real-Time hub, making that signal discoverable, routable, and consumable by your entire organization. For more information, see What is Fabric Activator? and Business events overview (Preview). |
| Activator rule actions: Copy job and Publish a business event (Preview) | Activator rules now support two new actions: Copy job (preview) for copying data between sources and destinations, and Publish a business event (preview) for triggering downstream processes that consume business events. For more information, see Configure actions for Activator rules. |
| Activator rules for OneLake items (Preview) | In the Fabric portal, you can create and manage Activator rules for OneLake items using Fabric Activator. These rules allow you to automate actions based on specific events or conditions related to OneLake items, such as when a file is created or deleted. You can also create rules based on the status of processes that interact with OneLake items, such as pipelines, Spark jobs, and notebooks. For more information, see Trigger Fabric items. |
| Advanced edit for data destination queries in Dataflow Gen2 (Preview) | You can now use advanced edit for data destination queries in Dataflow Gen2 to modify destination-side query logic directly in authoring. |
| ALTER COLUMN for Fabric Data Warehouse (Preview) | ALTER TABLE ... ALTER COLUMN (Preview) lets you make supported schema changes directly on warehouse tables with familiar T-SQL syntax. Supported scenarios complete as metadata-only updates without rewriting Parquet files. For more information, see ALTER TABLE (Transact-SQL) syntax for warehouse in Fabric. |
| ALTER DATABASE SET options in SQL database | ALTER DATABASE SET options in SQL database are now available as a preview feature. For more information, see ALTER DATABASE SET options (Transact-SQL) for SQL database in Fabric. Collation support and full-text indexing are also now available in SQL database as preview features. |
| Anomaly detection (Preview) | With a no-code interface, automatic model selection, and flexible alerts, tracking changes and unexpected events is easy with Anomaly detection in Real-Time Intelligence (Preview). For more information, see AI-Powered Real-Time Intelligence with Anomaly Detection (Preview). Billing for Anomaly Detector starts in December. |
| Apache Iceberg data in OneLake using Snowflake and shortcuts (Preview) | You can now consume Apache Iceberg-formatted data across Microsoft Fabric with no data movement or duplication, plus Snowflake added the ability to write Iceberg tables directly to OneLake. For more information, see Use Apache Iceberg tables with OneLake. |
| Approval activity in Data Factory pipelines (Preview) | The new Approval activity (Preview) in Fabric Data Factory pipelines pauses execution until a reviewer approves or rejects through Outlook or Teams, then routes pipeline flow dynamically based on the outcome (approved, rejected, or timed out). For more information, see Approval activity documentation. |
| Associate an identity for items (Preview) | Associated identities for items (Preview) let you associate a user, service principal, or managed identity with Fabric Lakehouses and Eventstreams via REST API, removing the dependency on the item owner. For more information, see Manage identities associated with Fabric items. |
| Automatic index compaction in SQL database | Automatic index compaction helps you reduce the consumption of storage space, disk I/O, memory, and improve workload performance without investing time and effort into index maintenance jobs. |
| Azure Key Vault references to authenticate to Fabric data connections (Preview) | You can now authenticate to Fabric data connections using Azure Key Vault stored secrets (preview). Azure Key Vault references enable secure and centralized secret management for your data connections. For more information, see Azure Key Vault references overview (preview) and get started at Configure Azure Key Vault references. |
| Azure Monitor to Fabric Eventhouse (Preview) | Azure Monitor to Fabric Eventhouse (Preview) lets you route VM telemetry via Azure Monitor Agent and Data Collection Rules into Eventhouse for schema-managed ingestion, ad hoc queries, time-series analytics, and activation. For more information, see Send virtual machine client data to Fabric and Azure Data Explorer (Preview). |
| Business Events in Real-Time Intelligence (Preview) | Business Events (Preview) capture critical business moments from User Data Functions and Notebooks and enable actions via Activator alerts, custom logic, workflows, AI models, Spark jobs, dataflows, and Power Automate. For more information, see Business events overview (Preview). |
| Business events persisted into Eventhouse (Preview) | Eventhouse is enabled by default when you create a business event in Real-Time hub. Each business event maps to a dedicated KQL table in your eventhouse database, and every published event is automatically ingested and retained. For more information, see Business events overview (Preview). |
| Bulk import and export items definition (Preview) | Bulk import and export APIs (Preview) enable workspace migration, CI/CD deployment, metadata backup, and template provisioning by exporting and importing Fabric item definitions in bulk via the REST API. For more information, see Items API reference. |
| Capacity overage (Preview) | Capacity overage (Preview) is an opt-in capability that automatically bills for excess capacity usage during compute spikes instead of throttling, with admin-configurable limits per 24-hour window. For more information, see Capacity overage in Microsoft Fabric. |
| Centralized data governance in the OneLake catalog (Preview) | A new centralized data governance experience in the OneLake catalog is in preview. Data owners can view aggregated insights on the items they created, consider improving their governance by taking recommended actions, and access more information along with all available tools in Fabric. |
| CI/CD Support for SQL analytics endpoint (Preview) | CI/CD support for SQL analytics endpoint (Preview) lets you manage SQL analytics endpoint definitions as DacFx database projects in Git alongside other Fabric items and deploy incremental schema changes through Fabric Deployment Pipelines. For more information, see Source control with Warehouse. |
| Code Interpreter Tool for Fabric Data Agent (Preview) | Code Interpreter Tool (Preview) in Fabric Data Agent runs Python directly inside agent workflows so the agent goes beyond database queries to perform statistical analysis, forecasting, cohort analysis, and rich Python visualization. For more information, see Fabric Data Agents. |
| Code-First Hyperparameter Tuning preview | In Fabric Data Science, FLAML is now integrated for hyperparameter tuning, currently a preview feature. Fabric's flaml.tune feature streamlines this process, offering a cost-effective and efficient approach to hyperparameter tuning. |
| Conditional activity retries in Data Factory pipelines (Preview) | Conditional activity retries (Preview) let you define exactly when a pipeline activity retries based on the error code, error message, or both with AND/OR logic, so long-running pipelines automatically recover from transient failures while fast-failing on errors that don't self-resolve. For more information, see Activity-level retry policy. |
| Configurable Data Retention in Fabric Data Warehouse (Preview) | Configurable data retention (Preview) in Fabric Data Warehouse lets you set the data history retention window for a warehouse between 1 and 120 days with a single T-SQL command. This retention period governs time travel, point-in-time clones, restore points, and warehouse snapshots. For more information, see Data retention in Fabric warehouse. |
| Confluent Schema Registry Support in Eventstream (Preview) | Eventstream's Confluent Cloud for Apache Kafka streaming connector now supports decoding data from topics associated with a data contract in Confluent Schema Registry, enabling seamless ingestion, preview, and routing of schema-encoded streaming data in Fabric Real-Time Intelligence. For more information, see Add Confluent Cloud for Apache Kafka source to an eventstream. |
| Data Factory in Fabric connection recency (Preview) | Connection recency (Preview) adds last-linked-to-items and last-credentials-used properties to connections for audit and lifecycle management. For more information, see Data source management. |
| Continuous ingestion from Azure Storage to Eventhouse (Preview) | You can now use continuous ingestion from Azure Storage to Eventhouse (preview) to automatically and efficiently ingest data from Azure Storage into Eventhouse. For more information, see Get data from Azure Storage. |
| Copilot for Dataflow Gen 2 Modern Get Data | With Copilot for the Modern Get Data experience in Dataflow Gen 2, you can ingest and transform data with natural language commands. For a walkthrough, see Blog: Copilot in Modern Get Data (MGD) for Dataflow Gen 2. |
| Copilot for Data Warehouse Chat preview | You now see a Copilot button in the ribbon that starts a chat with Copilot for acceleration with any data warehousing task. For more information, see How to: Use the Copilot chat pane for Fabric Data Warehouse. |
| Copilot for SQL analytics endpoint (Preview) | The Copilot for SQL analytics endpoint introduces Copilot capabilities for the SQL analytics endpoint, enabling users to generate and optimize SQL queries using natural language. For more information, see Copilot for SQL analytics endpoint. |
| Copilot in Fabric is available worldwide | Copilot in Fabric is now available to all customers, including Copilot for Power BI, Copilot for Data Factory, Copilot for Data Science & Data Engineering, and Copilot for Writing KQL Queries. Read more in our Overview of Copilot in Fabric. |
| Copilot context awareness for notebooks (Preview) | The updated Copilot for Data Engineering and Data Science (Preview) provides built-in context awareness of your workspace, attached lakehouse schemas, notebook structure, and execution environment. Copilot can generate multi-step code, refactor existing logic, summarize complex notebooks, and diagnose failures with Fix with Copilot. For more information, see Copilot for Data Engineering and Data Science. |
| Activator rules in Eventstream (Preview) | Configure and manage Activator rules directly in Eventstream (Preview) to embed alert creation, condition definition, and rule management. For more information, see Set alert on an eventstream with Activator destination. |
| Copy job support for change data capture (CDC) (Preview) | Change Data Capture (CDC) in Copy Job is a powerful capability in Data Factory Data pipelines that enables efficient and automated replication of changed data including inserted, updated, and deleted records from a source to a destination. |
| Create AI Auto-Description for Semantic Models (Preview) | Auto-Description for Semantic Models (Preview) uses Copilot to generate clear descriptions for semantic models from the model's metadata and structure, which model owners and contributors can review, edit, and regenerate directly from the semantic model details page in the OneLake catalog. For more information, see OneLake catalog overview. |
| Creator Agent for SQL and Eventhouse sources in Fabric Data Agent (Preview) | Creator Agent (Preview) is an AI-assisted creation experience that generates and refines Fabric Data Agent configurations through an interactive workflow, replacing manual setup with schema- and conversation-aware recommendations for agent instructions, data source guidance, and example queries; the preview focuses on SQL and Eventhouse scenarios. For more information, see Fabric Data Agents. |
| Custom authorization for API for GraphQL (Preview) | Authorizer User Data Functions for API for GraphQL (Preview) let you attach a User Data Function that runs before each GraphQL operation and evaluates request context such as tenant ID, role, or token claims to allow, restrict, or deny access based on your own policy logic. For more information, see API for GraphQL overview and Create a user data function. |
| Custom CA and mTLS support in Eventstream connectors (Preview) | Custom CA and mTLS support in Eventstream connectors (Preview) lets you specify custom certificate authority and client certificates stored in your own Azure Key Vault when configuring a source, so Eventstream connectors can connect to Kafka-based sources (Apache Kafka, AWS MSK, Confluent Cloud for Apache Kafka) and Confluent Schema Registry that aren't covered by the predefined trusted CA list or that require mTLS. For more information, see Eventstream sources overview. |
| Custom Live Pools for Fabric Data Engineering (Preview) | Custom Live Pools for Fabric Data Engineering (Preview) let workspace and capacity admins schedule ready-to-run Spark capacity tied to an Environment artifact, with activation windows, lifecycle and hydration monitoring through Monitoring Hub, intentional library-publishing semantics, and support for workspaces with Managed Private Endpoints. For more information, see Configure custom live pools in Microsoft Fabric. |
| Customer Managed Keys (CMK) for Eventhouse (Preview) | Customer Managed Keys (CMK) for Eventhouse (Preview) lets you bring your own Azure Key Vault key to encrypt Eventhouse data at rest, meeting strict regulatory requirements while keeping you in control of the key lifecycle. For more information, see Data encryption with customer-managed keys in Fabric Eventhouse. |
| Data Agent support for Eventhouse functions, materialized views, and shortcuts (Preview) | Data Agent now discovers and queries Eventhouse user-defined functions (UDFs), materialized views, and shortcut tables, so it calls validated KQL logic, uses pre-aggregated views for faster answers, and reaches data federated from sources outside the eventhouse. For more information, see Fabric Data Agents. |
| Data protection: DLP restrict access for structured data in OneLake (Preview) | Expanded DLP restrict access (Preview) covers all structured data in OneLake including SQL databases, KQL databases, and warehouses. For more information, see DLP restrict access. |
| Data protection: DSPM for AI for Fabric Copilots and data agents (Preview) | Data Security Posture Management (DSPM) for AI for Fabric (Preview) monitors Copilot and data agent interactions to detect sensitive information in AI prompts and responses, investigate risky AI behavior, and apply governance through Purview Audit and eDiscovery. For more information, see DSPM for AI in Fabric. |
| Data Warehouse Monitor (Preview) | Data Warehouse Monitor (Preview) (previously Query Activity) is a unified monitoring UI for Fabric Data Warehouse that shows active and historical queries side by side, supports cross-execution performance analysis powered by Query Insights, and lets you cancel long-running queries with one click. For more information, see Monitor T-SQL queries in Fabric Data Warehouse. |
| Dataflow Gen2 diagnostics download (Preview) | Dataflow Gen2 diagnostics download (Preview) lets you download run-level diagnostic packages with structured logs and execution details for faster root-cause analysis and support collaboration. For more information, see Monitor dataflows. |
| Dataflow Gen2 Public APIs (Preview) | Data Factory APIs enable users to automate and manage dataflows, including creation, management, scheduling, and monitoring. For more information, see Use public parameters in Dataflow Gen2 (preview). |
| Data Factory Adaptive performance tuning (Preview) | Adaptive Performance Tuning is designed to intelligently optimize data movement performance based on your configuration and runtime context. This preview feature makes tuning Data Factory simpler, safer, and more effective without requiring deep manual expertise or trial-and-error adjustments. To enable, see the adaptivePerformanceTuning property when using copy activity. |
| Data Factory On-premises data gateway manual update option (Preview) | The Gateway December 2025 release (version 3000.298) provides a new manual update option (in preview) via the Fabric portal or API.The November release serves as the baseline version for this feature, and you can start performing manual updates beginning in December. For more information, see Update an on-premises data gateway. |
| Data Factory MCP (Preview) | Data Factory MCP (Preview) enables AI assistants to create, test, and deploy Dataflow Gen2 through natural language without manual configuration. For more information, see Data Factory MCP on GitHub. |
| Data Factory SharePoint data source site picker (Preview) | The SharePoint Site Picker (Preview) replaces manual URL entry with a browsable, searchable dropdown for selecting SharePoint sites across Dataflow Gen2, Pipelines, Copy job, and Lakehouse shortcuts. For more information, see SharePoint Folder connector. |
| Data replication from Lakehouse with Delta Change Feed (Preview) | The Fabric Lakehouse Table connector provides changed data from a Fabric Lakehouse via Delta Change Data Feed (CDF), to supported destinations. For more information, see Data Replication from Fabric Lakehouse with Delta Change Data Feed (Preview). |
| dbt Job in Fabric Data Factory (Preview) | A dbt Job in Fabric Data Factory lets you author, schedule, and monitor dbt projects natively with serverless execution, integrated testing and documentation, and governance via Entra ID and SQL security policies. For more information, see dbt job in Microsoft Fabric (preview). |
| Delta column mapping in the SQL analytics endpoint | SQL analytics endpoint now supports Delta tables with column mapping enabled. For more information, see Delta column mapping. This feature is currently in preview. |
| DeltaFlow transformation in Fabric Real-Time Intelligence's Eventstream (preview) | DeltaFlow is a capability in Fabric Eventstream that transforms raw Change Data Capture (CDC) events into a flattened, analytics-ready format. Instead of working with deeply nested Debezium JSON payloads, DeltaFlow produces tabular rows that closely mirror the structure of the source database tables, enriched with metadata columns that describe each change. Eventstream supports DeltaFlow transformations, which are currently in preview. For more information, see DeltaFlow transformation and Blog: Business Events in Microsoft Fabric (Preview). |
| Digital twin builder (Preview) | Digital twin builder (preview) is an item within the Real-Time Intelligence workload. Digital twins create data-driven, real-time representations of entities. It's a data modeling item that creates digital representations of real-world environments, to optimize physical operations using data. For more information, see What is digital twin builder (preview)? |
| Dropped warehouse recovery (Preview) | Dropped warehouse recovery (Preview) lets Workspace Administrators restore accidentally dropped warehouses from the Workspace Recycle Bin within a configurable retention window (7 to 90 days), recovering table schemas, data, snapshots, permissions, views, and stored procedures. For more information, see Manage workspaces. |
| Enhanced conversation with Microsoft Fabric Copilot (Preview) | We're introducing improvements to AI functionalities in Microsoft Fabric, including a new way to store chat prompts and history, improved accuracy of responses, and better context knowledge retention. |
| Entity diagram in Eventhouse KQL database (Preview) | Entity diagram in Eventhouse KQL database (Preview) adds a visual entity diagram for exploring tables, relationships, data flow, and schema violations in Eventhouse KQL databases. For more information, see View an entity diagram in KQL database (preview). |
| Evaluate Power Query programmatically (Preview) | Evaluate Power Query Programmatically in Microsoft Fabric (Preview) introduces a public REST API to execute Power Query M scripts programmatically, enabling automation, integration with Spark and pipelines, and access to 100+ data sources with 90-second timeout for transformations. For more information, see Execute Query API Reference. |
| Evaluate your Fabric data agents with the Python SDK (Preview) | You can now use the Python SDK to programmatically evaluate Fabric data agents. For more information, see Consume a Fabric Data Agent in Microsoft Copilot Studio (preview). |
| Eventstream Business Events publisher (Preview) | Eventstream now publishes Business Events as a built-in destination, so you filter, aggregate, threshold, and emit governed, discoverable business signals from the Eventstream canvas with no code. |
| Eventstream connectors support for private network streaming (Preview) | Eventstream connectors now support secure private network streaming (Preview) by using Azure virtual network as a bridge to connect on-premises or private cloud data sources, enabling real-time data ingestion with enterprise-grade security through VPN, ExpressRoute, or private endpoints. For more information, see What is an eventstream? |
| Eventstream Derived Streams in Direct Ingestion mode (Preview) | You can ingest data from a Fabric Eventstream to Eventhouse seamlessly either from an eventstream or using Eventhouse Get Data Wizard. This capability is now being extended to support eventstream derived streams in direct ingestion mode. For more information, see Fabric Eventhouse now supports Eventstream Derived Streams in Direct Ingestion mode (preview). |
| Eventstream workspace monitoring (Preview) | Workspace monitoring for Eventstreams (Preview) automatically creates three Eventhouse tables (EventStreamNodeStatus, EventStreamMetrics, EventStreamErrorMetrics) for per-minute data volume, watermark delay, backlog, and error counts queryable with KQL. Existing Eventstreams need a republish to start emitting telemetry. For more information, see What is an eventstream? and Workspace monitoring overview. |
| Excel to Delta tables with shortcut transformations (Preview) | Excel to Delta tables with shortcut transformations (Preview) extends the zero-code shortcut transformations ingestion model to multi-sheet .xlsx and .xls workbooks, with automatic sheet discovery, wildcard sheet matching, combined-table or table-per-sheet output, schema drift handling, continuous sync, and sanitization of sheet names into valid Delta table names. For more information, see Shortcuts file transformations. |
| Extended Auto partition support in Copy job for Oracle, Fabric Lakehouse, and SAP HANA (Preview) | Extended Auto partition support in Copy job (Preview) automatically scales out reads and writes for Oracle, Fabric Lakehouse, and SAP HANA without predefining partition columns or custom sharding logic. For more information, see What is Copy job in Data Factory. |
| Extended IoT Hub source Eventstream connector (Preview) | Extended Azure IoT Hub source connector (Preview) in Fabric Eventstream preserves all event metadata, including system properties (prefixed with ___src__) and user-defined application properties, so you can join device identity with telemetry, compute ingestion latency, and route events by device. For more information, see Add an Azure IoT Hub source to an eventstream. |
| Extended SCD Type 2 support in Copy job for Fabric Data Warehouse (Preview) | Extended SCD Type 2 support in Copy job (Preview) brings native Slowly Changing Dimension Type 2 with effective dating and soft-delete handling to Fabric Data Warehouse destinations, giving you an audit-ready view of every record's lifecycle. For more information, see Change data capture (CDC) in Copy job. |
| Fabric Activator integration in Fabric User Data functions (Preview) | Fabric Activator integration with Fabric User Data functions (preview) means you can create functions to process events from any source, including Fabric events and OneLake events. For more information, see Trigger Fabric items. |
| Fabric Apps backend with the Rayfin SDK (Preview) | Rayfin (Preview) is an open-source SDK and CLI that lets developers and coding agents define a complete application backend in code—data models, APIs, identity, access policies, and business logic—and deploy it to Microsoft Fabric, where the resulting apps run as first-class items with built-in governance and direct access to OneLake. For more information, see Fabric Apps. |
| Fabric Capacity overview events in Real-Time Hub (Preview) | Fabric Capacity overview events in Real-Time Hub (Preview) adds real-time capacity summary and state events so you monitor health, detect throttling, and trigger actions using Activator, Eventstream, and dashboards. |
| Fabric Data Agents + Microsoft Copilot Studio (Preview) | The preview of the integration between Fabric data agents and Microsoft Copilot Studio is now available. For more information, see multi-agent orchestration in Microsoft Copilot Studio. |
| Fabric data agent integration with Azure AI Agent Service (Preview) | We're excited to launch the integration of data agents in Fabric with Azure AI Agent Service from Microsoft Foundry. To get started, see Consume a Fabric Data Agent in Microsoft Copilot Studio (preview). The Fabric data agent SDK is also available in preview. |
| Fabric data agent integration with Microsoft Copilot Studio (Preview) | Fabric data agent is available in preview and can be added as an agent to your custom setup in Microsoft Copilot Studio. For more information, see Fabric data agent integration with Microsoft Copilot Studio (preview). |
| Fabric data factory Upsert table action in the Lakehouse Connector (Preview) | The Upsert (Preview) table action in the Lakehouse Connector is in preview. |
| Fabric Connection inside Notebook (Preview) | Fabric Connection inside Notebook (Preview) enables you to create and manage cloud data source connections directly within notebooks. For more information, see Fabric connection inside Notebook. |
| Fabric Migration Assessment for Data Factory (Preview) | The new Fabric Migration Assessment for Data Factory (Preview) provides a built-in assessment and migration experience that evaluates ADF pipeline readiness and migrates supported pipelines into a Fabric workspace with connection mapping. For more information, see Migration planning for Azure Data Factory. |
| Fabric CLI in Azure DevOps (Preview) | The Fabric CLI in Azure DevOps (Preview) provides the Fabric CLI (fab) as a built-in Azure DevOps pipeline task for automating workspace and item management without manual tool installation. Get started with the Fabric CLI (generally available) at https://aka.ms/FabCLI. |
| Fabric IQ (Preview) workload | Fabric IQ (preview) is a new workload for unifying business semantics across data, models, and systems to power intelligent agents and decisions grounded in a live, holistic view of the business. For more information, see What is Fabric IQ (preview)? |
| Fabric Remote MCP (Preview) | Fabric Remote MCP (Preview) is a cloud-hosted Model Context Protocol server that lets AI agents perform real, authenticated operations against Fabric workspaces, items, search, permissions, connections, and OneLake through Microsoft Entra ID with full audit logging, with no local install. For more information, see Get started with Fabric Core MCP Server. |
| Fabric Runtime 2.0 (Preview) | Fabric Runtime 2.0 (Preview) includes Apache Spark 4.0, Delta Lake 4.0, Java 21, Scala 2.13, and Python 3.12 on Azure Linux 3.0. For more information, see Fabric Runtime 2.0 and Apache Spark runtimes overview. |
| Fabric Spark Applications Comparison (Preview) | The Spark Applications Comparison feature lets users select and compare up to four Spark application runs side by side. For more information, see Blog: Fabric Spark Applications Comparison. |
| Fabric Spark Diagnostic Emitter (Preview) | The Fabric Apache Spark Diagnostic Emitter (preview) allows Apache Spark users to collect logs, event logs, and metrics from their Spark applications and send them to various destinations, including Azure Event Hubs, Azure storage, and Azure log analytics. |
| Folder REST API (Preview) | You can now create and manage workspace folders in automation scenarios and integrate with other systems and tools. The Folder Rest API is now in preview. To get started, see Fabric REST API Folders. |
| Git developer experiences (Preview) | New Git developer experiences (Preview) add Branched Workspaces for formal source-to-feature workspace relationships, Selective Branching to branch out with only the items you need, and the ability to Compare Code Changes for rich diff review before committing or syncing. |
| High Concurrency Support for the Fabric Livy API (Preview) | High Concurrency (HC) sessions for the Fabric Livy API (Preview) let multiple Spark workloads share a single managed Spark session with isolated REPLs, server-side session packing via sessionTag, parallel statement execution, and built-in surfacing in the Monitoring Hub. Up to five REPLs run per underlying Livy session. For more information, see High concurrency mode for the Fabric Livy API and Get started with high concurrency for the Fabric Livy API. |
| Improved NL2SQL Engine for Fabric Data Agent (Preview) | Improved NL2SQL Engine (Preview) for Fabric Data Agent improves accuracy, transparency, and resilience when translating natural language to SQL across lakehouses, warehouses, and mirrored databases. It uses example queries to follow patterns, asks clarifying questions when intent is ambiguous, and surfaces structured diagnostics, available when you enable the Preview Fabric Data Agent Runtime. For more information, see Fabric Data Agents. |
| Inline code completion in Fabric notebooks (Preview) | Fabric notebooks with inline code completion helps users write code faster and with fewer errors. For more information, see Copilot for Data Engineering and Data Science (preview). |
| Invoke SSIS Package activity in Data Factory pipelines (Preview) | The Invoke SSIS Package activity in Data Factory pipelines is currently in preview and enables package execution from pipeline orchestration in Fabric. For more information, read Blog: Invoke SSIS Package Activity in Microsoft Fabric (preview). |
| Item History in Fabric Capacity Metrics App (Preview) | The Item History page in the Fabric Capacity Metrics App (Preview) provides a 30-day view of compute capacity consumption with interactive visuals and slicers for workspace and item-level analysis. For more information, see Understand the metrics app item history page (preview). |
| JobInsight Diagnostics Library (Preview) | JobInsight is a diagnostics library to analyze completed Spark applications via APIs for queries, jobs, stages, tasks, executors, and event logs. For more information, see JobInsight diagnostics library (preview). |
| Lakehouse auto-binding in Git (Preview) | Lakehouse auto-binding in Git (Preview) resolves lakehouse references automatically as notebooks move across Git-connected workspaces, making notebooks portable across development, test, and production environments. For more information, see Notebook source control and deployment. |
| Lakehouse support for git integration and deployment pipelines (Preview) | The Lakehouse now integrates with the lifecycle management capabilities in Microsoft Fabric, providing a standardized collaboration between all development team members throughout the product's life. Lakehouse Lifecycle management facilitates an effective product versioning and release process by continuously delivering features and bug fixes into multiple environments. |
| Lakehouse utility suite (Preview) | The Lakehouse Utility Suite now includes the Lakehouse maintenance activity (Preview), to automate routine Lakehouse upkeep tasks, and the Refresh SQL endpoint activity (Preview), to refresh the Lakehouse SQL analytics endpoint on demand or on a schedule within Fabric pipelines. |
| Livy REST API (Preview) | The Fabric Livy endpoint lets users submit and execute their Spark code on the Spark compute within a designated Fabric workspace, eliminating the need to create a Notebook or Spark Job Definition item. The Livy API offers the ability to customize the execution environment through its integration with the Environment. |
| Load Fabric OneLake Data in Excel | Easily load Fabric OneLake data into Excel with integrated OneLake catalog and modern Get Data experience (preview). For more information and steps to get started, see OneLake catalog and Get Data are integrated into Excel for Windows. |
| Manage built-in SQL database mirroring to OneLake (preview) | Built-in database mirroring to OneLake makes operational data immediately available for analytics and AI, with zero ETL. You can now selectively manage the tables mirrored into Fabric OneLake, and start or restart the Mirroring capability using the REST API. |
| Manage failure notifications from the monitoring hub in Fabric (Preview) | The Schedule failures page in the Fabric monitoring hub (Preview) lists every scheduled item with failure notifications configured and lets you add, edit, or remove recipients for any item from one place, without opening each item individually. For more information, see Schedule failures (preview) and Receive notifications for failed scheduled jobs. |
| Maven support in Fabric Environment (Preview) | Maven support in Fabric Environment (Preview) lets Scala and Java developers add libraries from Maven repositories to a Fabric Environment by uploading and managing pom.xml files directly, so transitive dependencies are resolved without manually downloading and uploading JARs. For more information, see Manage libraries in Fabric environments. |
| MAX_ROLLOVER_FILES in extended event sessions using Azure Storage blobs (preview) | For new extended event sessions in SQL database in Fabric, you can specify the MAX_ROLLOVER_FILES option to retain only the specified number of the most recent Azure Storage blobs. For more information, see event_file target. |
| Max vCores limit option in SQL databases (Preview) | The Max vCores limit option (Preview) lets you cap SQL database compute at 4 or 32 vCores to control costs and protect shared capacities from noisy-neighbor scenarios. For more information, see Control compute usage (preview). |
| Microsoft ADO.NET Driver for Fabric Data Engineering (Preview) | The Microsoft ADO.NET Driver for Fabric Data Engineering (Preview) lets .NET applications connect to Spark SQL in Fabric using standard ADO.NET patterns with Microsoft Entra ID authentication, connection pooling, and async prefetch via Livy APIs. For more information, see Microsoft ADO.NET driver for Fabric Data Engineering. |
| MCP remote servers for Activator and Eventhouse in Real-Time Intelligence (preview) | Fabric Real-Time Intelligence provides hosted Model Context Protocol (MCP) remote servers for Activator and Eventhouse, enabling AI assistants and agents to interact with RTI components through natural language. By using the Activator MCP server, AI assistants can create monitoring rules, manage alerts, and trigger actions in Fabric Activator. The Eventhouse remote MCP server allows AI agents to query, reason over, and act on real-time data stored in an RTI eventhouse by configuring a URL that points to the Eventhouse. For more information, see Get started with the Eventhouse remote MCP server and Get started with the Activator remote MCP server. This feature is currently in preview. |
| Microsoft Fabric Admin APIs | Fabric Admin APIs are designed to streamline administrative tasks. The initial set of Fabric Admin APIs is tailored to simplify the discovery of workspaces, Fabric items, and user access details. |
| Microsoft Fabric SKU estimator (Preview) | The Microsoft Fabric SKU estimator, now available in preview, is an enhanced version of the previously introduced Microsoft Fabric Capacity Calculator. For more information, see Introducing the Microsoft Fabric SKU estimator (preview) and Mastering SKU Estimations with the Microsoft Fabric SKU Estimator. |
| Microsoft ODBC Driver (Preview) | The Microsoft ODBC Driver for Fabric Data Engineering (Preview) enables .NET, Python, and other ODBC-compatible applications and BI tools to connect to Spark SQL in Fabric through Livy APIs with Microsoft Entra ID authentication, session reuse, and async prefetch. For more information, see Microsoft ODBC driver for Fabric Data Engineering. |
| Migration Assistant to SQL database (preview) | The new Migration Assistant for SQL database simplifies moving SQL Server on-premises workloads into Fabric. Designed for SQL developers, it imports schema via DACPACs, identifies compatibility issues, and provides clear, actionable guidance before migration. For more information, see Fabric Migration Assistant for SQL database (Preview) and Introducing the Migration Assistant to SQL database in Fabric (preview). |
| Mirrored Dremio catalog in OneLake (Preview) | Fabric can mirror Dremio Iceberg REST Catalog metadata to enable zero-copy access to Dremio-managed Iceberg tables in OneLake via shortcuts. Tables appear in OneLake within seconds after selection, with optional auto-include of newly added tables and SQL analytics endpoint querying across Fabric workloads. For more information, see Bring your Dremio data into OneLake (Preview) and Mirrored Dremio catalog. |
| Mirroring for Azure Database for MySQL (Preview) | Mirroring in Fabric provides an easy experience to avoid complex ETL (Extract, Transform, Load) processes and seamlessly integrate your existing Azure Database for MySQL data with the rest of your data in Fabric. For more information, see Mirroring Azure Database for MySQL. |
| Mirroring for Google BigQuery (Preview) | Mirroring in Fabric offers a simple way to avoid complex ETL (Extract, Transform, Load) processes and seamlessly integrate your existing Google BigQuery warehouse data with the rest of your data in Fabric. For more information, see Mirrored databases from Google BigQuery and Tutorial: Set up mirroring for Google BigQuery. |
| ML model endpoints (Preview) | ML models in Fabric can now serve real-time predictions from secure, scalable, and easy-to-use online endpoints. In addition to batch predictions in Spark, you can use endpoints to bring ML model predictions to other Fabric solutions and custom applications. For more information, see Automated machine learning in Fabric and Model endpoints in Fabric. |
| MSSQL extension for VS Code Fabric integration (Preview) | MSSQL extension for VS Code Fabric integration (Preview) adds support for connecting, running queries, and managing objects in SQL databases in Fabric directly from Visual Studio Code. Download the extension at MSSQL extension at marketplace.visualstudio.com. |
| Multiple-Schema Inferencing in Eventstream (Preview) | Multiple-schema inferencing in Eventstream lets you work with multiple data sources that emit varying schemas by inferring and managing multiple schemas simultaneously. For more information, see Enhancing Data Transformation Flexibility with Multiple-Schema Inferencing in Eventstream (Preview). |
| New SQL analytics endpoint metadata sync option (preview) | You can now opt-in to the new SQL analytics endpoint metadata sync option, a preview feature. The new option, which is currently opt-in only and currently applies only to new SQL analytics endpoints, is faster, provides a new manual metadata sync option, and provides time travel querying. |
| My queries in Dataflow Gen2 (Preview) | My queries (Preview) gives Dataflow Gen2 authors a personal query library: right-click any query and Add to My Queries to save its M code to a per-user folder in My workspace, then reuse it across dataflows from the Recents & My Queries module in the modern Get data experience. For more information, see My queries (Preview) in Dataflow Gen2. |
| Notebook Copilot inline code completion (Preview) | Now in preview, Copilot Inline Code Completion (Preview) is an AI feature that assists data scientists and engineers in writing Python code more quickly and easily. For more information, see Notebook Copilot inline code completion (Preview). |
| Notebook debug within vscode.dev (Preview) | You can now place breakpoints and debug your Notebook code with the Synapse VS Code - Remote extension in vscode.dev. This update first starts with the Fabric Runtime 1.3 (GA). |
| Observability for Fabric Data Agent in Microsoft Foundry (Preview) | Observability for Fabric Data Agent in Microsoft Foundry (Preview) surfaces telemetry for every call to a Fabric Data Agent used as a tool inside a Foundry agent, including latency, status, and error details, so agent builders can debug and monitor production agents. For more information, see Fabric Data Agents. |
| OneLake Catalog search API, MCP, and CLI tools (Preview) | OneLake Catalog search API and MCP tool (Preview) brings cross-workspace discovery to code with a single search request to locate matching items across your accessible estate based on catalog metadata and user permissions, with the capability also included as a built-in tool in the Fabric Core MCP server for AI agents. The Fabric CLI fab find command extends the search to the terminal so you can locate items across workspaces, filter or exclude by item type, and pipe results through JMESPath. For more information, see OneLake catalog. |
| OneLake item-size reporting (Preview) | OneLake item-size reporting (Preview) gives workspace administrators item-level insight into storage usage across OneLake, including system and soft-deleted data, with a unified report and on-demand refresh. For more information, see Get the size of OneLake items. |
| OneLake storage tiers and lifecycle management (Preview) | OneLake now supports hot, cool, and cold storage tiers, and lifecycle management policies automatically move files between tiers based on creation time, last modification, or last access, so you trade higher transaction and retrieval costs for lower storage cost on infrequently accessed data. |
| On-premises data gateway manual update (Preview) | Manual update for the on-premises data gateway is now in preview, allowing administrators to trigger gateway updates through the Fabric portal, API, or PowerShell scripts on their own schedule. For more information, see Update an on-premises data gateway. |
| Ontology (preview) item | Ontology (preview) in Fabric IQ lets you define entity types, relationships, properties, and other constraints to organize data according to your business vocabulary. For more information, see What is ontology (preview)? |
| OpenAI plugins for Eventhouse (Preview) | You can now use two powerful AI plugins for Eventhouse: AI Embed Text Plugin and AI Chat Completion Prompt Plugin. Connect Eventhouse data to OpenAI-powered applications for advanced analytics and AI scenarios. For more information, see ai_embed_text (preview) and ai_chat_completion (preview). |
| Plan (preview) item | Plan (preview) in Fabric IQ is a unified no-code platform for collaborative planning, reporting, analytics, data integration, and management. For more information, see What is plan (preview)? |
| Pagination support for Eventstream HTTP connector (Preview) | Pagination support for the Eventstream HTTP connector (Preview) lets you ingest paginated REST API responses through page-based or cursor-based pagination, eliminating custom orchestration when bringing activity logs, audit records, or operational events into Eventstream. |
| Partitioned compute for Dataflow Gen2 (Preview) | Partitioned compute is a capability of the Dataflow Gen2 engine that allows parts of your dataflow logic to run in parallel, reducing the time to complete its evaluations. For more information, see partitioned compute in Dataflow Gen2. |
| Pass Parameter Values to Fabric Items (Preview) | Activator enables you to automatically activate Fabric items like pipeline and notebook whenever certain data conditions are met. You can not only activate and execute Fabric items but also pass values to the parameters defined in your Fabric items. |
| Pre & postscripts for Fabric Data Warehouse Deployments (Preview) | Pre & postscripts for Fabric Data Warehouse deployments (Preview) run controlled SQL actions before and after warehouse deployments to validate prerequisites, prepare the environment, apply permissions, seed data, and perform post-deployment validation as part of CI/CD pipelines. For more information, see Source control with Warehouse. |
| Prebuilt Foundry Tools in Fabric preview | The preview of prebuilt Foundry Tools in Fabric is an integration with Foundry Tools, formerly known as Azure AI services. Prebuilt Foundry Tools allow for easy enhancement of data with prebuilt AI models without any prerequisites. Currently, prebuilt Foundry Tools are in preview and include support for the Microsoft Azure OpenAI Service, Azure Language in Foundry Tools, and Azure Translator in Foundry Tools. |
| Public API capabilities for Dataflow Gen2 dataflows in Fabric Data Factory (Preview) | This preview of the Dataflows Gen 2 public APIs enables users to create, update, and monitor their data workflows programmatically. The APIs support a wide range of operations including dataflows CRUD (Create, Read, Update, and Delete), scheduling, and monitoring, making it easier for users to manage their data integration processes. |
| Public parameter values to refresh a Dataflow Gen2 (Preview) | Learn more about the new public parameters capability for Dataflow Gen2 with CI/CD support, and the support for this new mode within the Dataflow refresh activity in Data Pipelines. |
| Real-Time Dashboards powered by AI (Preview) | The redesigned Real-Time Dashboard tile editor (Preview) is an AI-first experience where you choose a visual type and describe what you need in natural language, and Copilot generates the visualization for review without requiring KQL expertise. Every Copilot-generated iteration is preserved in a per-visual chat history, and you can switch seamlessly between AI-assisted and manual editing. For more information, see Add or edit a tile in a Real-Time Dashboard. |
| Real-Time Intelligence Cribl source (Preview) | Cribl source (Preview) enables real-time data to flow into Fabric Eventstream from diverse telemetry and log sources through Cribl Stream, including Syslog, Datadog Agent, Splunk, Open Telemetry, and edge-based sources, with simplified Kafka endpoint configuration in Real-Time Hub. For more information, see Add Cribl source to an eventstream (preview). |
| Recent data in Dataflow Gen2 (Preview) | Recent data in Dataflow Gen2 (Preview) provides quick access to your most frequently used data items in the Power Query ribbon and Modern Get Data, letting you jump directly to recently used tables, files, folders, and databases without navigation. For more information, see Get data experience and What is Dataflow Gen2. |
| REST APIs for connections and gateways (Preview) | REST APIs for connections and gateways are now in preview. These new APIs allow developers to programmatically manage and interact with connections and gateways within Fabric. |
| Resource Instance Rules for OneLake (Preview) | Resource Instance Rules for OneLake (Preview) let workspace admins allow inbound access to OneLake from explicitly trusted Azure resource instances by identity, complementing Private Link and IP firewall rules for service-to-service scenarios without requiring IP allowlists or private networking. For more information, see Manage inbound access to OneLake with Resource Instance Rules. |
| Resource Profiles in Fabric Data Engineering (Preview) | Resource Profiles in Fabric Data Engineering (Preview) are preconfigured Spark compute profiles—write-heavy, read-heavy for Spark, and read-heavy for Power BI—that you select per environment so Fabric applies workload-aware settings automatically without manual tuning of dozens of Spark properties. For more information, see Configure resource profile configurations. |
| Scalar user-defined functions (UDFs) | Scalar user-defined functions (UDFs) are now supported as a preview feature. For more information, see CREATE FUNCTION for Fabric Data Warehouse. |
| Schema Registry (Preview) | Event Schema Registry (preview) provides a contract-based way to define and validate event schemas in Fabric Eventstreams for type-safe, reliable real-time pipelines. For more information, see Schema Registry overview. |
| Secure MQTT broker and Eventstream connector with mTLS (Preview) | mTLS for the Eventstream MQTT source connector (Preview) lets you specify your own CA and client certificates stored in Azure Key Vault, so Eventstream can establish a mutually authenticated, encrypted connection with an MQTT broker for secure IoT data ingestion across untrusted networks. |
| Service principal support for Fabric data agents (Preview) | Fabric data agents now support service principal (SPN) authentication, so custom apps and Microsoft Foundry agents call the data agent API with an application identity instead of a delegated user token. The data agent inherits workspace permissions; KQL database support is coming soon. |
| Solace PubSub+ Connector | Seamlessly connect Fabric Eventstream with Solace PubSub+ (preview). For more information and steps to get started, see New Solace PubSub+ Connector: seamlessly connect Fabric Eventstream with Solace PubSub+ (preview). |
| Spark Connector for SQL databases (Preview) | Spark Connector for SQL databases lets Spark read and write to Azure SQL Database, Azure SQL Managed Instance, SQL Server on Azure VM, and Fabric SQL databases with built-in authentication and PySpark support. For more information, see Spark connector for SQL databases documentation. |
| SQL database data virtualization (Preview) | Data virtualization in SQL database enables querying external data stored in OneLake using T-SQL. With data virtualization syntax, you can execute Transact-SQL (T-SQL) queries on files that store data in common data formats in OneLake. You can combine this data with locally stored relational data by using joins. |
| SQL database support for Tenant level private links (Preview) | You can use tenant level private links to provide secure access for data traffic in Microsoft Fabric, including SQL database (in preview). For more information, see Set up and use private links and Blog: Tenant Level Private Link (Preview). |
| Stream Mirrored Database change feeds into Eventstreams (Preview) | Mirrored Database change feed connector for Eventstreams (Preview) streams Delta Change Data Feed (CDF) row-level inserts, updates, and deletes from a mirrored database directly into a Fabric Eventstream for low-latency, event-driven processing, with full source schema and change metadata fidelity and support for Azure SQL, Cosmos DB, Oracle, PostgreSQL, Snowflake, and Open Mirroring partners. For more information, see Extended capabilities in mirroring. |
| Synapse Data Explorer to Eventhouse migration tooling (Preview) | The next generation of Azure Synapse Data Explorer offering is evolving to become Eventhouse. To get started, see Migrate from Azure Synapse Data Explorer to Fabric Eventhouse (preview). |
| Time Series Visualization in Real-Time Dashboards (Preview) | Time Series Visualization (Preview) in Real-Time Dashboards adds dedicated capabilities for navigating, comparing, and customizing time-based data, including a legend search bar, pin-and-overlay for series comparison, multiple aligned-time panels, flexible Y-axis scaling, color assignment, and zoom with a time slider. |
| Time Travel using SQL analytics endpoint (Preview) | Time Travel using SQL analytics endpoint (Preview) extends the Fabric Data Warehouse time travel experience to the SQL analytics endpoint so you can query Lakehouse data exactly as it looked at any prior point in time by adding OPTION (FOR TIMESTAMP AS OF '...') once at the T-SQL statement level. For more information, see Time travel in Fabric Data Warehouse. |
| Upsert to delta table with Lakehouse connector (Preview) | We've added upsert support to the Lakehouse connector, allowing direct writes to Delta tables, in both Copy job and Copy activity within Pipeline. For more information, see Configure Lakehouse in a copy activity. |
| Warehouse custom SQL pools (preview) | Custom SQL pools (Preview) give workspace administrators fine-grained control over SQL compute resource allocation, enabling you to define pools, assign percentage-based compute, and route queries by application name. For more information, see Custom SQL pools. |
| Warehouse data clustering (Preview) | Data clustering is a technique used to organize and store data based on similarity. Data clustering improves query performance and reduces compute and storage access costs for queries by grouping similar records together. For more information and to get started, see Data clustering documentation and Use data clustering in Fabric Data Warehouse. |
| Warehouse IDENTITY columns (Preview) | IDENTITY columns automatically produce unique values for each new row, eliminating the need for manual key assignments and eliminating the risk of key duplication and key integrity issues. For more information and to get started, see IDENTITY columns and Use IDENTITY columns to create surrogate keys. You can also learn how to Migrate to IDENTITY columns in Fabric Data Warehouse. |
| Warehouse source control (Preview) | Using Source control with Warehouse (preview), you can manage development and deployment of versioned warehouse objects. You can use SQL Database Projects extension available inside of Visual Studio Code. For more information on warehouse source control, see CI/CD with Warehouses in Microsoft Fabric. |
| Workspace-level surge protection (Preview) | Workspace-level surge protection controls (Preview) enable per-workspace CU percentage limits over a rolling 24-hour period, automatic blocking of workspaces exceeding thresholds, and mission critical mode to exempt high-priority workspaces from surge protection rules. For more information, see Surge protection. |
| Workspace customer-managed keys for BYOK (Preview) | Workspace CMK for BYOK (Preview) enables workspace-level customer-managed key encryption in BYOK-enabled Fabric capacities, so you can use the same or separate Azure Key Vault keys without dedicated capacities. For more information, see Customer-managed keys for Fabric workspaces. |
| Workspace monitoring (Preview) | Workspace monitoring is a Microsoft Fabric database that collects data from a range of Fabric items in your workspace, and lets you access and analyze logs and metrics. For more information about this feature, see Announcing preview of workspace monitoring. |
| Workspace Monitoring for Copy job (Preview) | Workspace Monitoring for Copy job (Preview) streams Copy job execution logs to a Monitoring Eventhouse with per-activity metrics including throughput, data volume, error codes, and timing for centralized observability. For more information, see Workspace Monitoring for Copy job. |
| Zero-copy access to OneLake data in Azure Databricks (Preview) | OneLake catalog federation in Azure Databricks (Preview) enables Unity Catalog to query Fabric data stored in OneLake without copying, keeping OneLake as the source of truth while allowing Databricks compute to analyze tables through synchronized metadata and zero-copy data access. For more information, see Enable OneLake catalog federation. |
Generally available features
The following table lists the features of Microsoft Fabric that recently transitioned to general availability (GA).
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Fabric Git Integration – GitHub Enterprise Cloud with Data Residency (Generally Available) | Fabric Git Integration now supports GitHub Enterprise Cloud with data residency (Generally Available), so customers using GitHub Enterprise Cloud with Data Residency (ghe.com) can connect Fabric workspaces directly to their GitHub repositories while keeping repository content in the region of their EMU. For more information, see GitHub Enterprise Cloud with Data Residency support in Fabric Git integration. |
| June 2026 | New default AI functions (Generally Available) | AI functions in Fabric (Generally Available) add GPT-5 and Phi-4 model support, remove the synapseml-core package dependency for Python users, and provide richer Spark execution and usage statistics including input/output token counts for cost monitoring. For more information, see AI functions overview. |
| June 2026 | Data agents in Microsoft 365 Copilot (Generally Available) | Fabric Data Agents in Microsoft 365 Copilot (Generally Available) lets business users discover and chat with governed Fabric data sources directly inside Microsoft 365 Copilot, with admin-managed publishing and Entra ID-enforced data permissions. For more information, see Fabric Data Agents. |
| June 2026 | Real-Time Dashboards Live Refresh (Generally Available) | Live Refresh for Real-Time Dashboards (Generally Available) pushes new data to dashboard tiles as soon as it lands, replacing periodic refresh polling with always-current visualizations for operational monitoring and live analytics scenarios. For more information, see Real-Time Dashboards overview. |
| June 2026 | Business Events Capacity Consumption (Generally Available) | Business Events capacity consumption is now metered through the standard Fabric capacity model, so publishing, routing, and consuming business events bills against your existing capacity SKU with no separate license. For more information, see Microsoft Fabric Capacity Metrics app and Business events overview. |
| June 2026 | Eventstream streaming connectors for Apache Kafka and Azure Service Bus (Generally Available) | Eventstream Apache Kafka and Azure Service Bus streaming connectors (Generally Available) provide hardened reliability, broader authentication support (SASL_SSL, SASL_PLAINTEXT, Microsoft Entra), and production-ready throughput for ingesting Kafka topics and Service Bus queues into Fabric Eventstream. For more information, see Add Apache Kafka source to an eventstream and Add Azure Service Bus source to an eventstream. |
| June 2026 | CDC with SQL estates in Copy job (Generally Available) | Change data capture (CDC) for SQL estates in Copy job (Generally Available) brings native, low-latency replication of inserts, updates, and deletes from Azure SQL Database, SQL Server, and Azure SQL Managed Instance into Fabric destinations with no separate change tracking infrastructure. For more information, see Change data capture (CDC) in Copy job. |
| June 2026 | SharePoint and OneDrive shortcuts (Generally Available) | Create shortcuts to SharePoint or OneDrive locations, with support for workspace identity and service principal authentication. For more information, see Create a OneDrive or SharePoint shortcut and Workspace identity. |
| June 2026 | Fabric Graph (Generally Available) | Graph in Microsoft Fabric helps you model, visualize, and analyze complex relationships within your data. It's a scalable, enterprise-grade solution that turns disconnected data into AI-powered insights. By using graph, you can uncover hidden connections within your data and enhance decision-making capabilities. For more information, see Graph in Microsoft Fabric overview. |
| June 2026 | Migration Assistant direct connection (Generally Available) | You can now use the Migration Assistant to connect directly to the source warehouse for a migration to Fabric Data Warehouse. For more information, see Fabric Migration Assistant for Data Warehouse and Migrate by using the connection to the source system. |
| May 2026 | Edit Copy job via JSON payloads (Generally Available) | Edit Copy job via JSON payloads (Generally Available) lets advanced users programmatically define and update Copy job configurations beyond the visual authoring experience, so you can automate edits and apply bulk, consistent changes across environments. For more information, see What is Copy job in Data Factory. |
| May 2026 | Switch between full and incremental copy mode in Copy job (Generally Available) | Switch between full and incremental copy mode in Copy job (Generally Available) lets you flip an existing Copy job between full (batch) and incremental modes with a single action—no recreation required—so you can adapt as change-tracking sources come and go. Incremental mode always starts with an initial full copy before subsequent incremental runs. For more information, see Incremental copy in Copy job. |
| May 2026 | Warehouse connection string changes (Generally Available) | Connection behavior changes in Fabric Data Warehouse make connections deterministic when the InitialCatalog connection string value is ambiguous or missing, improving predictability for downstream auditing. For more information, see Fabric Data Warehouse connectivity and Operation list. |
| May 2026 | Create support tickets from within Microsoft Fabric (Generally Available) | Ticket creation from within Microsoft Fabric (Generally Available) lets eligible Fabric and Power BI users open support cases from the Help Pane, which collects session metadata such as tenant location and item identifiers and routes them to support engineers without leaving Fabric. Admins control availability through the Publish "Get Help" information tenant setting. For more information, see Create a Fabric and Power BI support ticket and Publish "Get Help" information. |
| May 2026 | Capacity metrics app enhancements and Chargeback app (Generally Available) | Capacity health page, timepoint summary, timepoint detail, and the Chargeback app are generally available. The health page surfaces utilization and throttling signals across all capacities per region, timepoint summary and detail break down compute usage by workspace, item, operation, and user, and the Chargeback app supports cost allocation by SKU and workload type. For more information, see Understand the Capacity Metrics App health page and What is Microsoft Fabric Chargeback app?. |
| May 2026 | OneLake catalog natively available in Microsoft Foundry (Generally Available) | The OneLake catalog is now natively available in Microsoft Foundry, so AI builders can browse and select OneLake data directly inside Foundry, evaluate ownership, endorsement, sensitivity, and location signals in place, and turn governed assets into knowledge sources without leaving the project. For more information, see OneLake for Microsoft Foundry. |
| May 2026 | OneLake security (Generally Available) and OneLake data access roles (Generally Available) | OneLake security and OneLake data access roles are generally available, providing fine-grained access control for OneLake data with role permissions and folder-, row-, and column-level security. New capabilities include a wizard-based role creation flow with RLS and CLS at creation time, inline RLS validation, and cross-region shortcut support. OneLake security is being enabled by default on all supported items by the end of May 2026. For more information, see Get started with OneLake security and OneLake data access roles. |
| May 2026 | Refresh SQL endpoint activity (Generally Available) | The Refresh SQL endpoint activity in Fabric Data Factory pipelines is now generally available, enabling you to refresh the Lakehouse SQL analytics endpoint on demand or on a schedule within your pipeline orchestration. For more information, see Refresh SQL endpoint activity documentation. |
| April 2026 | Tabbed multitasking and object explorer (Generally Available) | Tabbed multitasking and object explorer (Generally Available) deliver fully supported tabs for working across multiple Fabric items in parallel, plus an object explorer that browses items across open workspaces. The GA release adds right-click tab management, open-in-new-window, a resizable explorer pane, support for opening monitoring jobs as tabs, and accessibility improvements. For more information, see Multitask with tabs and object explorer. |
| April 2026 | Eventstream SQL operator (Generally Available) | Eventstream SQL operator is generally available, the production-ready code-first transformation operator in Fabric Eventstreams, with multiple destinations from a single operator (Eventhouse, Lakehouse, Activator, downstream Eventstreams), built-in per-output testing in the canvas, and event-time processing with configurable late-arrival and out-of-order thresholds. For more information, see Process events using a SQL operator. |
| April 2026 | Nested folders support in shortcut transformations (Generally Available) | Nested folders support in shortcut transformations (Generally Available) extends the zero-ETL shortcut transformations model to multi-level folder hierarchies with recursive change detection, automatic partition pickup, directory structure preservation, and a single Delta table output per top-level folder. The Include subfolders option is enabled by default for new and existing transformations. For more information, see Shortcuts file transformations. |
| April 2026 | Outbound access protection for Data Factory (Generally Available) | Workspace outbound access protection (OAP) for Data Factory (Generally Available) extends workspace-level outbound controls to Pipelines, Copy Job, Dataflows, and Mirrored Databases (including Mirrored SQL Database and Mirrored Snowflake), so Data Factory items in a protected workspace only connect to admin-allowed endpoints. For more information, see Workspace outbound access protection overview and Workspace outbound access protection for Data Factory. |
| April 2026 | Cross-workspace logging for MLflow in Microsoft Fabric (Generally Available) | Cross-workspace logging for MLflow (Generally Available) lets you log MLflow experiments, metrics, parameters, and registered models from any environment (Fabric notebook, Azure Databricks, Azure Machine Learning, local) to any Fabric workspace via the synapseml-mlflow package and the MLFLOW_TRACKING_URI environment variable, enabling dev/test/prod MLOps separation and train-where-data-lives, serve-from-a-separate-workspace patterns. For more information, see Cross-workspace logging for MLflow in Microsoft Fabric. |
| April 2026 | Fabric Local MCP (Generally Available) | Fabric Local MCP (Generally Available) is the open-source on-machine Model Context Protocol server that gives AI assistants grounded knowledge of Fabric's APIs, OneLake file operations, and Fabric item operations with integrated authentication, error handling, auto-retry, and telemetry, and works with VS Code GitHub Copilot, Cursor, Claude Desktop, and any MCP-compatible client. The Fabric Remote MCP server is also available as a preview feature. For more information, see Get started with Fabric Pro-Dev MCP Server. |
| April 2026 | Warehouse ALTER TABLE support in an explicit transaction (Generally available) | Supported ALTER TABLE Transact-SQL operations can now execute inside an explicit user-defined transaction in Fabric Data Warehouse. For more information, see Transactions in Fabric Data Warehouse. |
| April 2026 | Shortcut transformations (Generally Available) | Shortcut transformations automatically turn structured files into Delta tables as you bring data into or move it within OneLake, keeping data always in sync without the need for pipelines. For more information, see Shortcuts file transformations. |
For older general availability (GA) announcements, review the Microsoft Fabric What's New archive.
Community
This section summarizes new Microsoft Fabric community opportunities for prospective and current influencers and MVPs.
- Sign up for the Fabric Community Newsletter: visit Fabric Community News and select Subscribe from the Options menu.
- Join a local Fabric User Group or join a local event.
- The Microsoft Fabric Career Hub has everything you need on your certification journey.
- Vote for your favorite new product feature ideas at Microsoft Fabric Ideas.
- To learn about the Microsoft MVP Award and to find MVPs, see mvp.microsoft.com.
- Are you a student? Learn more about the Microsoft Learn Student Ambassadors program.
- Watch and subscribe to Microsoft Fabric videos on YouTube.
- Ask and answer questions in the Microsoft Fabric community.
- Join the Microsoft Fabric user panel to share real-world experiences and feedback with the Fabric and Power BI product teams through surveys and 1:1 meetings.
- Spread your Fabric knowledge, insights, and best practices with others. To learn more, see the Super User Program.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Data Days | In 2026, we're going bigger than ever. Data Days now expands beyond Fabric and Power BI and includes SQL and AI. The event features over 100 live sessions, more than 5 contests and challenges, and dozens of study groups and learning opportunities. You aren't going to want to miss this event. It all starts on June 15, 2026. Pre-register to get updates on Data Days activities, voucher offers, and more. |
| June 2026 | FabCon+SQLCon Europe 2026 | Europe's largest Microsoft Fabric and SQL event lands in Barcelona, September 28 – October 1, 2026. From 130 sessions and 4 keynotes to workshops, the expo, community spaces, and the Power BI DataViz World Championships, this is where the data community comes together. Register now and save €200 with code FABCMTY200. |
| May 2026 | Fabric Influencers Spotlight May 2026 | Fabric Influencers Spotlight May 2026 highlights blog posts, videos, and presentations from Microsoft MVPs and Fabric Super Users covering Power BI, Data Engineering, Data Science, Fabric platform administration and governance, and databases. |
| May 2026 | Fabric Jumpstart | Fabric Jumpstart is a catalog of Microsoft-sponsored and community-contributed accelerators, tutorials, and demos for Microsoft Fabric that you install with a single Python command (fabric_jumpstart.install) to deploy end-to-end scenarios such as a real-time healthcare billing system or a stateful streaming lakehouse without building from scratch. |
| April 2026 | Fabric Influencers Spotlight April 2026 | Fabric Influencers Spotlight April 2026 highlights blog posts, videos, and presentations from Microsoft MVPs and Fabric Super Users covering Data Warehouse, Power BI, Data Engineering, Data Integration, Fabric platform and governance, and Real-Time Intelligence. |
| April 2026 | Update on UX feedback from the Reddit thread | The Fabric design team responded to community UX feedback with fixes for item dropdown menus going off-screen and keyboard shortcut conflicts, with dark mode, "Open in new tab" from Object Explorer, and tab grouping planned. |
| April 2026 | Five reasons to attend SQLBits in the UK | Microsoft SQL is the headline sponsor of SQLBits 2026 at ICC Wales (22-25 April), with 30+ sessions, a full-day AI + SQL training, and a migration panel. Use code 10MICROSOFT for 10% off registration. |
For older updates, review the Microsoft Fabric What's New archive.
Power BI
Important
If you access Power BI on a web browser version older than Chrome 94, Microsoft Edge 94, Safari 16.4, Firefox 93, or equivalent, you need to upgrade your web browser to a newer version by August 31, 2024. Using an outdated browser version after this date can prevent you from accessing features in Power BI.
For updates to Power BI Desktop and the Power BI service, see What's new in Power BI?
Microsoft Fabric platform features
News and feature announcements about the Microsoft Fabric platform experience.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Fabric Skills for AI coding tools | Fabric Skills is an open-source, MIT-licensed library of Microsoft-authored AI skills that teach GitHub Copilot, Claude Code, Cursor, Windsurf, and other AI coding tools the right APIs, token audiences, query patterns, and end-to-end recipes for Fabric workloads. Install the full fabric-skills bundle or focused fabric-authoring, fabric-consumption, or fabric-operations bundles from the skills-for-fabric repository. |
| June 2026 | Fabric Git Integration – GitHub Enterprise Cloud with Data Residency (Generally Available) | Fabric Git Integration now supports GitHub Enterprise Cloud with data residency (Generally Available), so customers using GitHub Enterprise Cloud with Data Residency (ghe.com) can connect Fabric workspaces directly to their GitHub repositories while keeping repository content in the region of their EMU. For more information, see GitHub Enterprise Cloud with Data Residency support in Fabric Git integration. |
| May 2026 | Notebook export controls | Notebook Export Control is a tenant-level and workspace-level security setting that lets admins block notebook downloads and rich DataFrame export operations, closing a data exfiltration path for regulated organizations. For more information, see Notebook data export controls. |
| May 2026 | Create support tickets from within Microsoft Fabric (Generally Available) | Ticket creation from within Microsoft Fabric (Generally Available) lets eligible Fabric and Power BI users open support cases from the Help Pane, which collects session metadata such as tenant location and item identifiers and routes them to support engineers without leaving Fabric. Admins control availability through the Publish "Get Help" information tenant setting. For more information, see Create a Fabric and Power BI support ticket and Publish "Get Help" information. |
| May 2026 | Fabric Jumpstart | Fabric Jumpstart is a catalog of Microsoft-sponsored and community-contributed accelerators, tutorials, and demos for Microsoft Fabric that you install with a single Python command (fabric_jumpstart.install) to deploy end-to-end scenarios such as a real-time healthcare billing system or a stateful streaming lakehouse without building from scratch. |
| May 2026 | Capacity metrics app enhancements and Chargeback app (Generally Available) | Capacity health page, timepoint summary, timepoint detail, and the Chargeback app are generally available. The health page surfaces utilization and throttling signals across all capacities per region, timepoint summary and detail break down compute usage by workspace, item, operation, and user, and the Chargeback app supports cost allocation by SKU and workload type. For more information, see Understand the Capacity Metrics App health page and What is Microsoft Fabric Chargeback app?. |
| May 2026 | OneLake catalog natively available in Microsoft Foundry (Generally Available) | The OneLake catalog is now natively available in Microsoft Foundry, so AI builders can browse and select OneLake data directly inside Foundry, evaluate ownership, endorsement, sensitivity, and location signals in place, and turn governed assets into knowledge sources without leaving the project. For more information, see OneLake for Microsoft Foundry. |
| May 2026 | OneLake security (Generally Available) and OneLake data access roles (Generally Available) | OneLake security and OneLake data access roles are generally available, providing fine-grained access control for OneLake data with role permissions and folder-, row-, and column-level security. New capabilities include a wizard-based role creation flow with RLS and CLS at creation time, inline RLS validation, and cross-region shortcut support. OneLake security is being enabled by default on all supported items by the end of May 2026. For more information, see Get started with OneLake security and OneLake data access roles. |
| May 2026 | OneLake storage tiers and lifecycle management (Preview) | OneLake now supports hot, cool, and cold storage tiers, and lifecycle management policies automatically move files between tiers based on creation time, last modification, or last access, so you trade higher transaction and retrieval costs for lower storage cost on infrequently accessed data. |
| May 2026 | OneLake item-size reporting (Preview) | OneLake item-size reporting (Preview) gives workspace administrators item-level insight into storage usage across OneLake, including system and soft-deleted data, with a unified report and on-demand refresh. For more information, see Get the size of OneLake items. |
| April 2026 | Tabbed multitasking and object explorer (Generally Available) | Tabbed multitasking and object explorer (Generally Available) deliver fully supported tabs for working across multiple Fabric items in parallel, plus an object explorer that browses items across open workspaces. The GA release adds right-click tab management, open-in-new-window, a resizable explorer pane, support for opening monitoring jobs as tabs, and accessibility improvements. For more information, see Multitask with tabs and object explorer. |
| April 2026 | Create AI Auto-Description for Semantic Models (Preview) | Auto-Description for Semantic Models (Preview) uses Copilot to generate clear descriptions for semantic models from the model's metadata and structure, which model owners and contributors can review, edit, and regenerate directly from the semantic model details page in the OneLake catalog. For more information, see OneLake catalog overview. |
| April 2026 | Manage failure notifications from the monitoring hub in Fabric (Preview) | The Schedule failures page in the Fabric monitoring hub (Preview) lists every scheduled item with failure notifications configured and lets you add, edit, or remove recipients for any item from one place, without opening each item individually. For more information, see Schedule failures (preview) and Receive notifications for failed scheduled jobs. |
| April 2026 | Resource instance rules for OneLake (Preview) | Resource instance rules for OneLake (Preview) let workspace admins allow inbound access to OneLake from explicitly trusted Azure resource instances by identity, complementing Private Link and IP firewall rules for service-to-service scenarios without requiring IP allowlists or private networking. For more information, see Manage inbound access to OneLake with Resource Instance Rules. |
| April 2026 | Outbound access protection for Data Factory (Generally Available) | Workspace outbound access protection (OAP) for Data Factory (Generally Available) extends workspace-level outbound controls to Pipelines, Copy Job, Dataflows, and Mirrored Databases (including Mirrored SQL Database and Mirrored Snowflake), so Data Factory items in a protected workspace only connect to admin-allowed endpoints. Data Agent and Eventstreams gain OAP support in preview. For more information, see Workspace outbound access protection overview and Workspace outbound access protection for Data Factory. |
| April 2026 | Fabric Local MCP (Generally Available) and Fabric Remote MCP (Preview) | Agentic Fabric: How MCP is turning your data platform into an AI-native operating system introduces two Model Context Protocol entry points to Fabric: Local MCP, an open-source on-machine server that grounds AI assistants in Fabric APIs, OneLake file operations, and Fabric item operations, and Remote MCP, a cloud-hosted server that lets agents perform authenticated workspace, item, search, permission, and connection operations through Microsoft Entra ID with full audit logging. The Fabric Remote MCP server is also available as a preview feature. For more information, see Get started with Fabric Pro-Dev MCP Server and Get started with Fabric Core MCP Server. |
| April 2026 | Shortcut transformations (Generally Available) | Shortcut transformations automatically turn structured files into Delta tables as you bring data into or move it within OneLake, keeping data always in sync without the need for pipelines. For more information, see Shortcuts file transformations. |
| April 2026 | Workspace customer-managed keys for BYOK (Preview) | Workspace customer-managed keys (CMK) for BYOK (Preview) enables workspace-level CMK in BYOK-enabled Fabric capacities, so you can use the same or separate Azure Key Vault keys for Power BI semantic models and other Fabric items without provisioning dedicated capacities. For more information, see Customer-managed keys for Fabric workspaces. |
| April 2026 | Associate an identity for items (Preview) | Associated identities for items (Preview) let you associate a user, service principal, or managed identity with Fabric Lakehouses and Eventstreams via REST API, removing the dependency on the item owner. For more information, see Manage identities associated with Fabric items. |
For older updates, review the Microsoft Fabric What's New archive.
Continuous Integration/Continuous Delivery (CI/CD) in Microsoft Fabric
This section includes guidance and documentation updates on development process, tools, source control, and versioning in the Microsoft Fabric workspace.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Fabric Git Integration – GitHub Enterprise Cloud with Data Residency (Generally Available) | Fabric Git Integration now supports GitHub Enterprise Cloud with data residency (Generally Available), so customers using GitHub Enterprise Cloud with Data Residency (ghe.com) can connect Fabric workspaces directly to their GitHub repositories while keeping repository content in the region of their EMU. For more information, see GitHub Enterprise Cloud with Data Residency support in Fabric Git integration. |
| June 2026 | CI/CD Support for SQL analytics endpoint (Preview) | CI/CD support for SQL analytics endpoint (Preview) lets you manage SQL analytics endpoint definitions as DacFx database projects in Git alongside other Fabric items and deploy incremental schema changes through Fabric Deployment Pipelines. |
| June 2026 | Pre & postscripts for Fabric Data Warehouse Deployments (Preview) | Pre & postscripts for Fabric Data Warehouse deployments (Preview) run controlled SQL actions before and after warehouse deployments to validate prerequisites, prepare the environment, apply permissions, seed data, and perform post-deployment validation as part of CI/CD pipelines. |
| March 2026 | Fabric variable libraries in Dataflow Gen2 with CI/CD (Generally available) | Fabric variable libraries offer a centralized way to manage configuration values across Microsoft Fabric workloads. With integration in Dataflow Gen2, you can reference these variables directly in your dataflow, enabling dynamic behavior across environments and simplifying CI/CD workflows. For more information, see variable libraries in Dataflow Gen2. |
For older updates, review the Microsoft Fabric What's New archive.
Data Factory in Microsoft Fabric
This section summarizes recent new features and capabilities of Data Factory in Microsoft Fabric. Follow issues and feedback through the Data Factory Community Forum.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | CDC with SQL estates in Copy job (Generally Available) | Change data capture (CDC) for SQL estates in Copy job (Generally Available) brings native, low-latency replication of inserts, updates, and deletes from Azure SQL Database, SQL Server, and Azure SQL Managed Instance into Fabric destinations. For more information, see Change data capture (CDC) in Copy job. |
| June 2026 | Extended SCD Type 2 support in Copy job for Fabric Data Warehouse (Preview) | Extended SCD Type 2 support in Copy job (Preview) brings native Slowly Changing Dimension Type 2 with effective dating and soft-delete handling to Fabric Data Warehouse destinations, giving you an audit-ready view of every record's lifecycle. For more information, see Change data capture (CDC) in Copy job. |
| June 2026 | Extended Auto partition support in Copy job for Oracle, Fabric Lakehouse, and SAP HANA (Preview) | Extended Auto partition support in Copy job (Preview) automatically scales out reads and writes for Oracle, Fabric Lakehouse, and SAP HANA without predefining partition columns or custom sharding logic. For more information, see What is Copy job in Data Factory. |
| June 2026 | Conditional activity retries in Data Factory pipelines (Preview) | Conditional activity retries (Preview) let you define exactly when a pipeline activity retries based on the error code, error message, or both with AND/OR logic, so long-running pipelines automatically recover from transient failures while fast-failing on errors that don't self-resolve. For more information, see Activity-level retry policy. |
| May 2026 | Edit Copy job via JSON payloads (Generally Available) | Edit Copy job via JSON payloads (Generally Available) lets advanced users programmatically define and update Copy job configurations beyond the visual authoring experience, so you can automate edits and apply bulk, consistent changes across environments. For more information, see What is Copy job in Data Factory. |
| May 2026 | Switch between full and incremental copy mode in Copy job (Generally Available) | Switch between full and incremental copy mode in Copy job (Generally Available) lets you flip an existing Copy job between full (batch) and incremental modes with a single action—no recreation required—so you can adapt as change-tracking sources come and go. Incremental mode always starts with an initial full copy before subsequent incremental runs. For more information, see Incremental copy in Copy job. |
| May 2026 | On-premises data gateway May 2026 release | The May 2026 on-premises data gateway release brings the gateway up to date with the May 2026 Power BI Desktop release and introduces Admin consent for gateway diagnostics with integrated Dataflow Gen2 diagnostics (Preview), so tenant administrators control whether gateway diagnostic data is collected and gateway logs appear alongside dataflow execution details for unified troubleshooting. For more information, see Update an on-premises data gateway. |
| May 2026 | My queries in Dataflow Gen2 (Preview) | My queries (Preview) gives Dataflow Gen2 authors a personal query library: right-click any query and Add to My Queries to save its M code to a per-user folder in My workspace, then reuse it across dataflows from the Recents & My Queries module in the modern Get data experience. For more information, see My queries (Preview) in Dataflow Gen2. |
| May 2026 | Approval activity in Data Factory pipelines (Preview) | The new Approval activity (Preview) in Fabric Data Factory pipelines pauses execution until a reviewer approves or rejects through Outlook or Teams, then routes pipeline flow dynamically based on the outcome (approved, rejected, or timed out). For more information, see Approval activity documentation. |
| May 2026 | Business Workflow Management in Fabric Pipelines | Business Workflow Management in Fabric Data Factory pipelines enables end-to-end business workflows that orchestrate work across systems with built-in process tracking and status management. For more information, see Business Workflow Management documentation. |
| May 2026 | Copy job for SAP via ABAP Add-On (Preview) | Copy job for SAP via ABAP Add-On (Preview) extends SAP connectors in Fabric with a Microsoft Data Integration ABAP Add-On that extracts data from SAP S/4HANA and ECC systems, supporting full and incremental copy modes for tables, views, and CDS views. For more information, see Copy job for SAP via ABAP Add-On tutorial. |
| May 2026 | Pipeline canvas node experience updates | The refreshed pipeline canvas has new node designs for improved readability and usability when building and monitoring Fabric Data Factory pipelines. For more information, see Pipeline canvas reference. |
| May 2026 | Refresh Materialized Lake View activity in Data Factory pipelines | The new Refresh Materialized Lake View activity in Fabric Data Factory pipelines lets you refresh materialized lake views on demand or on a schedule within your pipeline orchestration. For more information, see Refresh Materialized Lake View activity documentation. |
| May 2026 | Refresh SQL endpoint activity (Generally Available) | The Refresh SQL endpoint activity in Fabric Data Factory pipelines is now generally available, enabling you to refresh the Lakehouse SQL analytics endpoint on demand or on a schedule within your pipeline orchestration. For more information, see Refresh SQL endpoint activity documentation. |
| May 2026 | Workspace Identity in Fabric Airflow Jobs | Workspace Identity in Fabric Airflow Jobs lets you authenticate Apache Airflow jobs using workspace identity, enabling secure access to Fabric resources without managing individual credentials. For more information, see Use Workspace Identity in Fabric Airflow Jobs. |
| April 2026 | On-premises data gateway April 2026 release | The April 2026 on-premises data gateway release (version 3000.314) brings the gateway up to date with the April 2026 Power BI Desktop release and reinforces on-premises Data Gateway Auto-Update (Admin Triggered) as Generally Available, so gateway administrators can manually trigger gateway updates on demand to align with maintenance windows. For more information, see Update an on-premises data gateway. |
| April 2026 | Outbound access protection for Data Factory (Generally Available) | Workspace outbound access protection (OAP) for Data Factory (Generally Available) extends workspace-level outbound controls to Pipelines, Copy Job, Dataflows, and Mirrored Databases (including Mirrored SQL Database and Mirrored Snowflake), so Data Factory items in a protected workspace only connect to admin-allowed endpoints. Data Agent and Eventstreams gain OAP support in preview. For more information, see Workspace outbound access protection overview and Workspace outbound access protection for Data Factory. |
For older updates, review the Microsoft Fabric What's New archive.
Data Factory in Microsoft Fabric samples and guidance
| Month | Feature | Learn more |
|---|---|---|
| May 2026 | Pipelines are evolving beyond ETL | This perspective post discusses how pipelines in Fabric are evolving from traditional ETL into end-to-end workflows that orchestrate work across systems, pause for human input through the new Approval activity (Preview), and observe long-running runs with workflow-aware monitoring. For more information, see Business Process Management in pipelines documentation. |
Fabric Apps (Preview)
This section summarizes recent announcements about Fabric Apps, which helps you build applications on the Microsoft Fabric platform.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Rayfin SDK and CLI (Preview) | Rayfin (Preview) is an open-source SDK and CLI that developers and coding agents use to define a complete application backend in code—data models, APIs, identity, access policies, and business logic—and deploy it to Microsoft Fabric. The resulting apps run as first-class items with built-in governance and direct access to OneLake. For more information, see Fabric Apps, the Rayfin GitHub repo, and the Rayfin docs. |
| May 2026 | Fabric Apps | Fabric Apps (preview) helps you build data-driven applications on Microsoft Fabric by combining data models, generated APIs, authentication, and hosting in one development workflow. You define your data models in TypeScript, and Fabric Apps uses them to generate the backend pieces your app needs. Fabric Apps supports TypeScript as the language for data models, client code, and application logic. |
Fabric Data Engineering
This section summarizes recent new features and capabilities of the Data Engineering workload in Microsoft Fabric.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Time Travel using SQL analytics endpoint (Preview) | Time Travel using SQL analytics endpoint (Preview) extends the Fabric Data Warehouse time travel experience to the SQL analytics endpoint so you can query Lakehouse data exactly as it looked at any prior point in time by adding OPTION (FOR TIMESTAMP AS OF '...') once at the T-SQL statement level. |
| May 2026 | Incremental Liquid Clustering in Fabric Runtime 2.0 | Incremental Liquid Clustering in Fabric Runtime 2.0 is a new OPTIMIZE algorithm that processes only unclustered, small, or deletion-vector-heavy files instead of rewriting entire 100-GB groups, with Auto Reclustering to keep layout quality high—delivering up to 8.9x faster clustering and constant-time cost that scales with new data, not table size. Enabled by default with no configuration changes; use OPTIMIZE table FULL for a full recluster. For more information, see Liquid Clustering for Delta tables in Microsoft Fabric. |
| May 2026 | Custom Live Pools for Fabric Data Engineering (Preview) | Custom Live Pools for Fabric Data Engineering (Preview) let workspace and capacity admins schedule ready-to-run Spark capacity tied to an Environment artifact, with activation windows, lifecycle and hydration monitoring through Monitoring Hub, intentional library-publishing semantics, and support for workspaces with Managed Private Endpoints. For more information, see Configure custom live pools in Microsoft Fabric. |
| May 2026 | Excel to Delta tables with shortcut transformations (Preview) | Excel to Delta tables with shortcut transformations (Preview) extends the zero-code shortcut transformations ingestion model to multi-sheet .xlsx and .xls workbooks, with automatic sheet discovery, wildcard sheet matching, combined-table or table-per-sheet output, schema drift handling, continuous sync, and sanitization of sheet names into valid Delta table names. For more information, see Shortcuts file transformations. |
| May 2026 | New SQL analytics endpoint metadata sync option (preview) | You can now opt-in to the new SQL analytics endpoint metadata sync option, a preview feature. The new option, which is currently opt-in only and currently applies only to new SQL analytics endpoints, is faster, provides a new manual metadata sync option, and provides time travel querying. |
| May 2026 | Resource Profiles in Fabric Data Engineering (Preview) | Resource Profiles in Fabric Data Engineering (Preview) are preconfigured Spark compute profiles—write-heavy, read-heavy for Spark, and read-heavy for Power BI—that you select per environment so Fabric applies workload-aware settings automatically without manual tuning of dozens of Spark properties. For more information, see Configure resource profile configurations. |
| May 2026 | Custom authorization for API for GraphQL (Preview) | Authorizer User Data Functions for API for GraphQL (Preview) let you attach a User Data Function that runs before each GraphQL operation and evaluates request context such as tenant ID, role, or token claims to allow, restrict, or deny access based on your own policy logic. For more information, see API for GraphQL overview and Create a user data function. |
| May 2026 | High Concurrency Support for the Fabric Livy API (Preview) | High Concurrency (HC) sessions for the Fabric Livy API (Preview) let multiple Spark workloads share a single managed Spark session with isolated REPLs, server-side session packing via sessionTag, parallel statement execution, and built-in surfacing in the Monitoring Hub. For more information, see High concurrency mode for the Fabric Livy API and Get started with high concurrency for the Fabric Livy API. |
| April 2026 | Maven support in Fabric Environment (Preview) | Maven support in Fabric Environment (Preview) lets Scala and Java developers add libraries from Maven repositories to a Fabric Environment by uploading and managing pom.xml files directly, so transitive dependencies are resolved without manually downloading and uploading JARs. For more information, see Manage libraries in Fabric environments. |
| April 2026 | Nested folders support in shortcut transformations (Generally Available) | Nested folders support in shortcut transformations (Generally Available) extends the zero-ETL shortcut transformations model to multi-level folder hierarchies with recursive change detection, automatic partition pickup, directory structure preservation, and a single Delta table output per top-level folder. The Include subfolders option is enabled by default for new and existing transformations. For more information, see Shortcuts file transformations. |
| April 2026 | Microsoft ADO.NET Driver for Fabric Data Engineering (Preview) | The Microsoft ADO.NET Driver for Fabric Data Engineering (Preview) lets .NET applications connect to Spark SQL in Fabric using standard ADO.NET patterns (DbConnection, DbCommand, DbDataReader) with Microsoft Entra ID authentication, connection pooling, session reuse, and async prefetch via Livy APIs. For more information, see Microsoft ADO.NET driver for Fabric Data Engineering. |
For older updates, review the Microsoft Fabric What's New archive.
Fabric Data Engineering samples and guidance
| Month | Feature | Learn more |
|---|---|---|
| May 2026 | Understand your storage with OneLake item-size reporting | OneLake item-size reporting walks through how workspace admins use the new Storage report under Workspace settings > OneLake to break down item-level storage across visible, hidden system, and soft-deleted data, sort and search items by size, and identify which Lakehouses, Warehouses, notebooks, and other items are driving OneLake consumption. For more information, see Get the size of OneLake items. |
| April 2026 | Bring Dremio data into OneLake with the Mirrored Dremio catalog (Preview) | The Mirrored Dremio catalog (Preview) connects Fabric to a Dremio Iceberg REST Catalog endpoint with credential vending and creates OneLake shortcuts to selected tables for zero-copy, no-ETL access across Fabric workloads, with optional auto-include for new tables and ongoing sync. For more information, see Mirrored Dremio catalog. |
Fabric Data Science
This section summarizes recent improvements and features for Data Science in Microsoft Fabric.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | New default AI functions (Generally Available) | AI functions in Fabric (Generally Available) add GPT-5 and Phi-4 model support, remove the synapseml-core package dependency for Python users, and provide richer Spark execution and usage statistics including input/output token counts for cost monitoring. For more information, see AI functions overview. |
| May 2026 | High Concurrency Support for the Fabric Livy API (Preview) | High Concurrency (HC) sessions for the Fabric Livy API (Preview) let multiple Spark workloads share a single managed Spark session with isolated REPLs, server-side session packing via sessionTag, parallel statement execution, and built-in surfacing in the Monitoring Hub. For more information, see High concurrency mode for the Fabric Livy API and Get started with high concurrency for the Fabric Livy API. |
| May 2026 | Data Agent support for Eventhouse functions, materialized views, and shortcuts (Preview) | Data Agent now discovers and queries Eventhouse user-defined functions (UDFs), materialized views, and shortcut tables, so it calls validated KQL logic, uses pre-aggregated views for faster answers, and reaches data federated from sources outside the eventhouse. For more information, see Fabric Data Agents. |
| April 2026 | Cross-workspace logging for MLflow in Microsoft Fabric (Generally Available) | Cross-workspace logging for MLflow (Generally Available) lets you log MLflow experiments, metrics, parameters, and registered models from any environment (Fabric notebook, Azure Databricks, Azure Machine Learning, local) to any Fabric workspace via the synapseml-mlflow package and the MLFLOW_TRACKING_URI environment variable, enabling dev/test/prod MLOps separation and train-where-data-lives, serve-from-a-separate-workspace patterns. For more information, see Cross-workspace logging for MLflow in Microsoft Fabric. |
For older updates, review the Microsoft Fabric What's New archive.
Fabric Data Science samples and guidance
| Month | Feature | Learn more |
|---|---|---|
| February 2026 | Enrich Power BI reports with ML | Machine learning can enrich Power BI reports. Learn how with this end-to-end pattern that uses Semantic Link to query a governed semantic model, trains a churn-prediction model with Fabric ML, applies batch and real-time scoring, and surfaces predictions in Power BI through Dataflow Gen2. |
Cosmos DB in Microsoft Fabric
This section summarizes recent improvements and features for Cosmos DB in Microsoft Fabric.
| Month | Feature | Learn more |
|---|---|---|
| November 2025 | Cosmos DB in Microsoft Fabric (GA) | Cosmos DB in Microsoft Fabric is now generally available to all users. Since its announcement at Microsoft Build 2025, Microsoft added several new capabilities, including vector indexing and search. For more information, see Announcing Cosmos DB in Microsoft Fabric. To get started, see Quickstart: Create a Cosmos DB database in Microsoft Fabric (preview). |
SQL database in Microsoft Fabric
This section summarizes recent improvements and features for SQL databases in Microsoft Fabric.
| Month | Feature | Learn more |
|---|---|---|
| May 2026 | MAX_ROLLOVER_FILES in extended event sessions using Azure Storage blobs (preview) | For new extended event sessions, you can specify the MAX_ROLLOVER_FILES option to retain only the specified number of the most recent Azure Storage blobs. For more information, see event_file target. |
| April 2026 | Migration Assistant to SQL database (preview) | The new Migration Assistant for SQL database simplifies moving SQL Server on-premises workloads into Fabric. Designed for SQL developers, it imports schema via DACPACs, identifies compatibility issues, and provides clear, actionable guidance before migration. For more information, see Fabric Migration Assistant for SQL database (Preview) and Introducing the Migration Assistant to SQL database in Fabric (preview). |
For older updates, review the Microsoft Fabric What's New archive.
SQL database in Microsoft Fabric samples and guidance
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Use case for SQL databases in AI applications | The latest in a series of articles, read the many ways you can use a Fabric SQL database in AI applications, including for the native vector data type and vector scalar functions, RAG-ready architecture, LangChain and Semantic Kernel, and Fabric platform integration. |
| June 2026 | Power BI in-report alerts with translytical task flows | A translytical task flow pattern for in-report alerts uses a SQL database in Fabric to store notifications, a User Data Function to write them back, and a Lakehouse shortcut to a Direct Lake semantic model to surface them as a per-report badge and detail panel inside Power BI reports—replacing email distribution lists with in-report alerts that any author can wire up. For more information, see Set up alerts directly in Power BI reports using translytical task flows. |
| May 2026 | Database schema in Git for SQL database in Fabric | SQL database in Microsoft Fabric ships with built-in Git integration, so you commit schema as SQL files alongside application code in GitHub or Azure DevOps, review changes through pull requests, and deploy them to Fabric workspaces with the same workflow you already use for app code. For more information, see SQL database in Microsoft Fabric source control integration and SQL Database Projects documentation. |
Fabric Data Warehouse
This section summarizes recent improvements and features for Fabric Data Warehouse.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Concatenation and fuzzy matching functions | Fabric Data Warehouse expands its T‑SQL language with String matching functions including (Levenshtein distance, Jaro–Winkler distance, Levenshtein similarity, and Jaro–Winkler similarity), Concatenation ( || and ||= ), and Unicode construction (UNISTR). |
| June 2026 | CI/CD Support for SQL analytics endpoint (Preview) | CI/CD support for SQL analytics endpoint (Preview) lets you manage SQL analytics endpoint definitions as DacFx database projects in Git alongside other Fabric items and deploy incremental schema changes through Fabric Deployment Pipelines. For more information, see Source control with Warehouse. |
| June 2026 | Pre & postscripts for Fabric Data Warehouse Deployments (Preview) | Pre & postscripts for Fabric Data Warehouse deployments (Preview) run controlled SQL actions before and after warehouse deployments to validate prerequisites, prepare the environment, apply permissions, seed data, and perform post-deployment validation as part of CI/CD pipelines. For more information, see Source control with Warehouse. |
| June 2026 | Time Travel using SQL analytics endpoint (Preview) | Time Travel using SQL analytics endpoint (Preview) extends the Fabric Data Warehouse time travel experience to the SQL analytics endpoint so you can query Lakehouse data exactly as it looked at any prior point in time by adding OPTION (FOR TIMESTAMP AS OF '...') once at the T-SQL statement level. For more information, see Time travel in Fabric Data Warehouse. |
| June 2026 | Migration Assistant direct connection (Generally Available) | You can now use the Migration Assistant to connect directly to the source warehouse for a migration to Fabric Data Warehouse. For more information, see Fabric Migration Assistant for Data Warehouse and Migrate by using the connection to the source system. |
| May 2026 | New SQL analytics endpoint metadata sync option (preview) | You can now opt-in to the new SQL analytics endpoint metadata sync option, a preview feature. The new option, which is currently opt-in only and currently applies only to new SQL analytics endpoints, is faster, provides a new manual metadata sync option, and provides time travel querying. |
| May 2026 | Warehouse connection string changes (Generally Available) | Connection string behavior changes in Fabric Data Warehouse make connections deterministic when the Initial Catalog value is ambiguous or missing, improving predictability for downstream auditing. For more information, see Fabric Data Warehouse connectivity and Operation list. |
| May 2026 | Fabric Data Warehouse upgrades in SSMS 22.5 | SSMS 22.5 Fabric Data Warehouse upgrades add direct create, rename, and delete actions for Fabric warehouses from Object Explorer, let you generate a SQL database project from a live warehouse connection, and remove the workspace-level permission requirement so item-level permissions are sufficient to connect. For more information, see Install SQL Server Management Studio (SSMS). |
| May 2026 | Data Warehouse Monitor (Preview) | Data Warehouse Monitor (Preview) (previously Query Activity) is a unified monitoring UI for Fabric Data Warehouse that shows active and historical queries side by side, supports cross-execution performance analysis powered by Query Insights, and lets you cancel long-running queries with one click. For more information, see Monitor T-SQL queries in Fabric Data Warehouse. |
| May 2026 | ALTER COLUMN for Fabric Data Warehouse (Preview) | ALTER TABLE ... ALTER COLUMN (Preview) lets you make supported schema changes directly on warehouse tables using familiar T-SQL syntax. Supported scenarios complete as metadata-only updates without rewriting Parquet files. For more information, see ALTER TABLE (Transact-SQL) syntax for warehouse in Fabric. |
| May 2026 | Configurable Data Retention in Fabric Data Warehouse (Preview) | Configurable data retention (Preview) in Fabric Data Warehouse lets you set the data history retention window for a warehouse between 1 and 120 days with a single T-SQL command. This governs time travel, point-in-time clones, restore points, and warehouse snapshots. For more information, see Data retention in Fabric Data Warehouse. |
| April 2026 | Dropped warehouse recovery (Preview) | Dropped warehouse recovery (Preview) lets Workspace Administrators restore accidentally dropped warehouses from the Workspace Recycle Bin within a configurable retention window (7 to 90 days), recovering table schemas, data, snapshots, permissions, views, and stored procedures. For more information, see Manage workspaces. |
| April 2026 | ALTER TABLE support in an explicit transaction (Generally available) | Supported ALTER TABLE Transact-SQL operations can now execute inside an explicit user-defined transaction in Fabric Data Warehouse. For more information, see Transactions in Fabric Data Warehouse. |
For older updates, review the Microsoft Fabric What's New archive.
Fabric Data Warehouse samples and guidance
| Month | Feature | Learn more |
|---|---|---|
| May 2026 | Fabric Data Warehouse as the modernization path for Synapse Dedicated SQL Pool | Modernizing Synapse Dedicated SQL Pool to Fabric Data Warehouse outlines how Fabric Data Warehouse addresses common Synapse pain points (pause/resume scaling, hard concurrency limits, manual hash/replicated distribution, architectural sprawl) with non-disruptive online scaling, autonomous workload management, intelligent data distribution, OneLake-based open Delta storage, true data virtualization, near real-time mirroring, and the AI-assisted Migration Assistant. For more information, see What is Fabric Data Warehouse? and Scale your Fabric capacity. |
| May 2026 | Direct Lake on SQL with Fabric Data Warehouse | Direct Lake on SQL with Fabric Data Warehouse kicks off a multi-part Direct Lake best practices series, explaining how Direct Lake transcodes Delta Parquet data on demand, why Fabric Data Warehouse automates file consolidation, statistics, V-Order, and clustering for guardrail-compliant Direct Lake workloads, and how to optimize cardinality, star schema, relationships, and cache state. For more information, see Direct Lake overview and Fabric Data Warehouse architecture. |
| April 2026 | Unstructured text processing with AI functions in Fabric Data Warehouse | Working with unstructured text using AI functions (Preview) demonstrates how to extract information, classify text, analyze sentiment, fix grammar, summarize, translate, and apply custom prompts directly in T-SQL. For more information, see Use AI functions in Fabric Data Warehouse (preview). |
Fabric Mirroring
This section summarizes recent improvements and features for Mirroring in Microsoft Fabric.
| Month | Feature | Learn more |
|---|---|---|
| March 2026 | Mirroring from SAP databases (Generally Available) | You can continuously replicate your SAP data directly into Fabric's OneLake. Once in Fabric, you can take advantage of powerful capabilities for business intelligence, AI, data engineering, data science, and data sharing. For more information, see Mirroring SAP. |
For older updates, review the Microsoft Fabric What's New archive.
Fabric Mirroring samples and guidance
| Month | Feature | Learn more |
|---|---|---|
| November 2025 | Mirroring: Uploading your CSVs | Open Mirroring improvements for CSV files remove the primary key requirement and automatically insert updates into existing mirrored tables in OneLake, with optional primary key and __rowMarker__ for advanced change tracking. For more information, see Open mirroring. |
Real-Time Intelligence in Microsoft Fabric
This section summarizes recent improvements and features for Real-Time Intelligence in Microsoft Fabric.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Real-Time Dashboards Live Refresh (Generally Available) | Live Refresh for Real-Time Dashboards (Generally Available) pushes new data to dashboard tiles as soon as it lands, replacing periodic refresh polling with always-current visualizations. For more information, see Real-Time Dashboards overview. |
| June 2026 | Eventstream streaming connectors for Apache Kafka and Azure Service Bus (Generally Available) | Eventstream Apache Kafka and Azure Service Bus streaming connectors (Generally Available) provide hardened reliability, broader authentication support (SASL_SSL, SASL_PLAINTEXT, Microsoft Entra), and production-ready throughput for ingesting Kafka topics and Service Bus queues into Fabric Eventstream. For more information, see Add Apache Kafka source to an eventstream and Add Azure Service Bus source to an eventstream. |
| June 2026 | Business Events Capacity Consumption (Generally Available) | Business Events capacity consumption is now metered through the standard Fabric capacity model, so publishing, routing, and consuming business events bills against your existing capacity SKU with no separate license. For more information, see Business events overview and Microsoft Fabric Capacity Metrics app. |
| June 2026 | Real-Time Dashboards powered by AI (Preview) | The redesigned Real-Time Dashboard tile editor (Preview) is an AI-first experience where you choose a visual type and describe what you need in natural language, and Copilot generates the visualization for review without requiring KQL expertise. For more information, see Add or edit a tile in a Real-Time Dashboard. |
| June 2026 | Time Series Visualization in Real-Time Dashboards (Preview) | Time Series Visualization (Preview) in Real-Time Dashboards adds dedicated capabilities for navigating, comparing, and customizing time-based data, including a legend search bar, pin-and-overlay for series comparison, multiple aligned-time panels, flexible Y-axis scaling, color assignment, and zoom with a time slider. |
| June 2026 | Secure MQTT broker and Eventstream connector with mTLS (Preview) | mTLS for the Eventstream MQTT source connector (Preview) lets you specify your own CA and client certificates stored in Azure Key Vault, so Eventstream can establish a mutually authenticated, encrypted connection with an MQTT broker for secure IoT data ingestion across untrusted networks. |
| June 2026 | Extended IoT Hub source Eventstream connector (Preview) | Extended Azure IoT Hub source connector (Preview) in Fabric Eventstream preserves all event metadata, including system properties (prefixed with ___src__) and user-defined application properties, so you can join device identity with telemetry, compute ingestion latency, and route events by device. For more information, see Add an Azure IoT Hub source to an eventstream. |
| June 2026 | Pagination support for Eventstream HTTP connector (Preview) | Pagination support for the Eventstream HTTP connector (Preview) lets you ingest paginated REST API responses through page-based or cursor-based pagination, eliminating custom orchestration when bringing activity logs, audit records, or operational events into Eventstream. |
| June 2026 | Fabric Graph (Generally Available) | Graph in Microsoft Fabric helps you model, visualize, and analyze complex relationships within your data. It's a scalable, enterprise-grade solution that turns disconnected data into AI-powered insights. By using graph, you can uncover hidden connections within your data and enhance decision-making capabilities. For more information, see Graph in Microsoft Fabric overview. |
| June 2026 | Activator as business events publisher (Preview) | Activator provides a no-code way to publish business events in Microsoft Fabric. When Activator detects that a condition is met in your data, it can emit a structured business event into Real-Time hub, making that signal discoverable, routable, and consumable by your entire organization. For more information, see What is Fabric Activator? and Business events overview (Preview). |
| June 2026 | Business events persisted into Eventhouse (Preview) | Eventhouse is enabled by default when you create a business event in Real-Time hub. Each business event maps to a dedicated KQL table in your eventhouse database, and every published event is automatically ingested and retained. For more information, see Business events overview (Preview). |
| June 2026 | Activator rule actions: Copy job and Publish a business event (Preview) | Activator rules now support two new actions: Copy job (preview) for copying data between sources and destinations, and Publish a business event (preview) for triggering downstream processes that consume business events. For more information, see Configure actions for Activator rules. |
| June 2026 | Activator rules for OneLake items (Preview) | In the Fabric portal, you can create and manage Activator rules for OneLake items using Fabric Activator. These rules allow you to automate actions based on specific events or conditions related to OneLake items, such as when a file is created or deleted. You can also create rules based on the status of processes that interact with OneLake items, such as pipelines, Spark jobs, and notebooks. For more information, see Trigger Fabric items. |
| May 2026 | Eventstream Business Events publisher (Preview) | Eventstream now publishes Business Events as a built-in destination, so you filter, aggregate, threshold, and emit governed, discoverable business signals from the Eventstream canvas with no code. |
| April 2026 | Stream Mirrored Database change feeds into Eventstreams (Preview) | Mirrored Database change feed connector for Eventstreams (Preview) streams Delta Change Data Feed (CDF) row-level inserts, updates, and deletes from a mirrored database directly into a Fabric Eventstream for low-latency, event-driven processing, with full source schema and change metadata fidelity and support for Azure SQL, Cosmos DB, Oracle, PostgreSQL, Snowflake, and Open Mirroring partners. For more information, see Extended capabilities in mirroring. |
| April 2026 | Custom CA and mTLS support in Eventstream connectors (Preview) | Custom CA and mTLS support in Eventstream connectors (Preview) lets you specify custom certificate authority and client certificates stored in your own Azure Key Vault when configuring a source, so Eventstream connectors can connect to Kafka-based sources (Apache Kafka, AWS MSK, Confluent Cloud for Apache Kafka) and Confluent Schema Registry that aren't covered by the predefined trusted CA list or that require mTLS. For more information, see Eventstream sources overview. |
| April 2026 | Eventstream SQL operator (Generally Available) | Eventstream SQL operator is generally available, the production-ready code-first transformation operator in Fabric Eventstreams, with multiple destinations from a single operator (Eventhouse, Lakehouse, Activator, downstream Eventstreams), built-in per-output testing in the canvas, and event-time processing with configurable late-arrival and out-of-order thresholds. For more information, see Process events using a SQL operator. |
| April 2026 | Customer Managed Keys (CMK) for Eventhouse (Preview) | Customer Managed Keys (CMK) for Eventhouse (Preview) lets you bring your own Azure Key Vault key to encrypt Eventhouse data at rest, meeting strict regulatory requirements while keeping you in control of the key lifecycle. For more information, see Data encryption with customer-managed keys in Fabric Eventhouse. |
| April 2026 | Stream SQL Change Events to Fabric Eventstream | Stream SQL Change Events (CES) to Fabric Eventstream shows how SQL Server 2025, Azure SQL Database, and Azure SQL Managed Instance push CloudEvents-formatted insert, update, and delete events directly into Eventstream's custom endpoint over AMQP or Kafka with Microsoft Entra (managed identity), SAS token, or SAS key authentication, eliminating CDC polling latency and intermediate change tables. For more information, see Stream SQL change events to Fabric Eventstream by using Change Event Streaming. |
| April 2026 | Eventstream workspace monitoring (Preview) | Workspace monitoring for Eventstreams (Preview) automatically creates three Eventhouse tables (EventStreamNodeStatus, EventStreamMetrics, EventStreamErrorMetrics) for per-minute data volume, watermark delay, backlog, and error counts queryable with KQL. Existing Eventstreams need a republish to start emitting telemetry. For more information, see What is an eventstream? and Workspace monitoring overview. |
For older updates, review the Microsoft Fabric What's New archive.
Real-Time Intelligence samples and guidance
Tip
Use the Real-Time Intelligence end-to-end sample solution to automatically create a collection of sample components.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Analyze Business Events in Eventhouse and Real-Time Dashboards (Preview) | Analyzing Business Events in Eventhouse and Real-Time Dashboards (Preview) walks through how every published business event lands in a dedicated KQL table in Eventhouse by default, so you can run historical KQL queries, correlate signals across event types, and connect a Real-Time Dashboard for live operational monitoring without extra pipelines. For more information, see Business events overview (Preview). |
| June 2026 | Activator as a Business Event Publisher (Preview) | Using Activator as a Business Event publisher (Preview) shows how Activator rules in Power BI reports, Real-Time Dashboards, KQL queries, and Fabric Warehouse SQL queries emit structured, governed business events into Real-Time hub, so signals are persistent, discoverable, and consumable across your organization. For more information, see What is Fabric Activator? and Business events overview (Preview). |
| March 2026 | Copilot for Real-Time Dashboard visuals (Preview) | Copilot for Real-Time Dashboard visuals (Preview) lets dashboard editors describe desired insights in natural language, and Copilot generates the KQL query, returns data, and suggests the best visualization format. For more information, see Copilot-assisted real-time data exploration. |
Fabric IQ (preview)
This section summarizes recent improvements and features for the new Fabric IQ (preview) workload.
| Month | Feature | Learn more |
|---|---|---|
| June 2026 | Data agents in Microsoft 365 Copilot (Generally Available) | Fabric Data Agents in Microsoft 365 Copilot (Generally Available) lets business users discover and chat with governed Fabric data sources directly inside Microsoft 365 Copilot, with admin-managed publishing and Entra ID-enforced data permissions. For more information, see Fabric Data Agents. |
| June 2026 | Observability for Fabric Data Agent in Microsoft Foundry (Preview) | Observability for Fabric Data Agent in Microsoft Foundry (Preview) surfaces telemetry for every call to a Fabric Data Agent used as a tool inside a Foundry agent, including latency, status, and error details, so agent builders can debug and monitor production agents. For more information, see Fabric Data Agents. |
| June 2026 | Creator Agent for SQL and Eventhouse sources in Fabric Data Agent (Preview) | Creator Agent (Preview) is an AI-assisted creation experience that generates and refines Fabric Data Agent configurations through an interactive workflow, replacing manual setup with schema- and conversation-aware recommendations for agent instructions, data source guidance, and example queries; the preview focuses on SQL and Eventhouse scenarios. For more information, see Fabric Data Agents. |
| June 2026 | Improved NL2SQL Engine for Fabric Data Agent (Preview) | Improved NL2SQL Engine (Preview) for Fabric Data Agent improves accuracy, transparency, and resilience when translating natural language to SQL across lakehouses, warehouses, and mirrored databases. It uses example queries to follow patterns, asks clarifying questions when intent is ambiguous, and surfaces structured diagnostics. For more information, see Fabric Data Agents. |
| June 2026 | Code Interpreter Tool for Fabric Data Agent (Preview) | Code Interpreter Tool (Preview) in Fabric Data Agent runs Python directly inside agent workflows so the agent goes beyond database queries to perform statistical analysis, forecasting, cohort analysis, and rich Python visualization. For more information, see Fabric Data Agents. |
| May 2026 | Service principal support for Fabric data agents (Preview) | Fabric data agents now support service principal (SPN) authentication, so custom apps and Microsoft Foundry agents call the data agent API with an application identity instead of a delegated user token. The data agent inherits workspace permissions; KQL database support is coming soon. |
For older updates, see the Microsoft Fabric What's New archive.