Azure Releases
Microsoft Azure updates and Terraform AzureRM provider. New features, breaking changes, security advisories and deprecations - each summarised in plain English and updated continuously.
Tracking 421 Azure releases · Updated
- Azure Updates aiazuredatabrickspreviewengineer ·
Azure Updates AI-powered discovery in preview
Microsoft is previewing an AI-powered feature for Azure Updates that uses the Release Communications MCP server to discover new Azure products and features. This aims to help customers meet cloud investment needs by keeping them informed of relevant changes. The feature is currently available to all Azure customers for non-production use and testing.
announcement feature - Azure Updates aiazuredatabrickspreviewengineer ·
Azure Databricks workspace-wide Copilot Studio integration now in preview
Azure Databricks is introducing a workspace-wide integration with Microsoft Copilot Studio, enabling AI-powered discovery of Azure Updates through the Microsoft Release Communications MCP server. This feature is currently in public preview and available for non-production use and testing by all Azure customers. It aims to help users stay informed about Azure product and feature updates to manage cloud investments effectively.
feature announcement - Azure Updates aiazuregaengineerazure-cosmos-db ·
LangChain integration for Azure Cosmos DB now generally available
A new Python package for LangChain and LangGraph is now generally available, integrating directly with Azure Cosmos DB. This enables faster development of AI and agent applications by allowing Cosmos DB to be used for vector and hybrid search, semantic caching, and chat history. The package is available for production use, facilitating quicker integration for developers building these types of applications.
feature announcement - Azure Updates aiazurepreviewengineerazure-cosmos-db ·
Azure Cosmos DB adds spherical quantization for vector search (preview)
Azure Cosmos DB is introducing spherical quantization in public preview, a new vector compression technique that enhances vector indexing speed and accuracy. This feature aims to improve search recall and precision for large-scale vector operations. It is now available for users to test and adopt.
feature announcement - Azure Updates aiazuregaengineerazure-functions ·
Azure Functions Durable Task Scheduler Consumption SKU GA
The Durable Task Scheduler Consumption SKU for Azure Functions is now generally available, enabling users to run durable workflows and AI agent orchestrations with pay-per-use pricing. This eliminates the need for storage management, capacity planning, and paying for idle resources. It is now generally available for production use.
feature announcement - Azure Updates aiazuregaengineer ·
Microsoft Agent Framework 1.0 now generally available
Microsoft Agent Framework has reached version 1.0 with stable APIs and long-term support for .NET and Python developers. This release enables multi-agent orchestration, multi-provider model support, and cross-runtime interoperability. It is now generally available for production use.
feature announcement - Azure Updates aiazurepreviewengineer ·
Foundry Agent Service Adds Memory in Public Preview
Foundry Agent Service now includes a managed long-term memory capability, currently in public preview. This feature integrates natively with Microsoft Agent Framework and LangGraph, eliminating the need for external databases. This simplifies the process for developers building agent-based applications on Azure.
feature announcement - Azure Updates aiazuregaengineer ·
Foundry Agent Service Supports Bring-Your-Own-Model Scenarios
Foundry Agent Service now allows agents to use external AI models, including those behind enterprise gateways like Azure API Management. This feature enhances flexibility by enabling organizations to leverage their existing AI investments and choose models best suited for their needs. The capability is now generally available, offering broader integration options for AI deployments.
feature announcement - Azure Updates aiazuredeprecationengineer ·
Prompt Flow to be retired April 2027 for Microsoft Agent Framework
Prompt Flow in Azure Machine Learning and Microsoft Foundry will be retired on April 20, 2027, replaced by the Microsoft Agent Framework. This consolidation aims to provide a unified platform with enhanced capabilities for orchestration. Existing users should plan to migrate their workloads to the new framework before the retirement date.
deprecation announcement - Azure Updates aiazuregaengineer ·
Foundry Toolkit for VS Code generally available
The Foundry Toolkit for VS Code, previously known as the AI Toolkit, is now generally available. It offers VS Code tooling for Microsoft Agent Framework, enabling agent creation, local testing, debugging, and deployment. This release benefits developers building agents within the Microsoft ecosystem, providing a more integrated and efficient workflow directly in their preferred IDE.
feature announcement - Azure Updates aiazuredatabricksgaengineer ·
Azure Databricks adds Anthropic Claude Opus 4.7 via AI Model Serving
Azure Databricks now offers Anthropic's Claude Opus 4.7 through its AI Model Serving. This integration brings Anthropic's most capable hybrid reasoning model to Azure Databricks users, improving performance on complex extraction and agentic reasoning tasks. The model is now generally available for all Azure Databricks customers.
feature announcement - Azure Updates aiazuregaengineer ·
Azure Red Hat OpenShift adds NVIDIA H100/H200 GPU support
Azure Red Hat OpenShift (ARO) now generally available supports NVIDIA H100 and H200 GPU-based Azure Virtual Machine SKUs. This enables customers to run large-scale AI, machine learning, and high-performance computing workloads. The update enhances ARO's capabilities for demanding computational tasks.
feature announcement - Azure Updates aiazuredeprecationengineer ·
Microsoft to Retire Eight Azure AI Language Features
Microsoft will retire eight Azure AI Language features on March 31, 2029, including Key Phrase Extraction and Sentiment Analysis. This change affects users of these specific Azure AI Language services, prompting a need to migrate to newer capabilities. The retirement is part of ongoing service evolution and users should plan their transition before the deadline.
deprecation
About Azure release tracking on ReleaseBytes
Azure updates land across the Azure Updates feed, service blogs and the Terraform AzureRM provider changelog. ReleaseBytes pulls them into one place, summarises each announcement in plain English, and flags retirements, breaking changes and security advisories — the items that actually require action — so you can filter past the routine noise.
Frequently asked questions
How often are Azure release notes updated on ReleaseBytes? ›
Continuously. ReleaseBytes monitors the official Azure release channels around the clock and publishes a plain-English summary of each announcement shortly after it lands.
What kinds of Azure changes does ReleaseBytes track? ›
New features, enhancements, bug fixes, security advisories, breaking changes, deprecations and end-of-life announcements. Every item is tagged by type so you can filter to just the changes that need action.
How can I get alerts for new Azure releases? ›
Set up a free email or Slack alert filtered to Azure, subscribe to the weekly digest, or follow the RSS feed. Teams can also install the ReleaseBytes GitHub App or connect via MCP.
Where does the Azure release data come from? ›
From the official sources: Microsoft Azure updates and Terraform AzureRM provider. Every item links back to the original vendor announcement.