Releasesby Glyph[Launched] Generally Available: Toolboxes in Microsoft Foundry
Microsoft Foundry brings Toolboxes to GA, providing a unified endpoint to invoke the most-used tools in prompt agents and callable outside of prompt agents. Today every agent team hand-assembles its own tool list to solve the same tasks, with no shared ar
Also shipped Jun 30, 2026Microsoft in June 2026 →
Analytical search (Public Preview)
Analytical search, an orchestration capability in Cortex Agents that enables analytical queries over large document collections, is now in Public Preview. Traditional retrieval-augmented generation (RAG) approaches retrieve a small number of passages, which limits their ability to answer questions requiring comprehensive analysis, such as counts...
MATCH_RECOGNIZE is now in Beta
Databricks SQL and Databricks Runtime now support the SQL:2016 MATCH_RECOGNIZE clause for row pattern recognition. Enable the Match Recognize preview in workspace preview settings . See MATCH_RECOGNIZE clause .
Lakehouse Real-Time (Beta)
Lakehouse Real-Time (Lakehouse//RT) is now in Beta . Lakehouse//RT is a serverless SQL warehouse type built for low-latency, high-concurrency workloads, delivering sub-second SQL read queries against Unity Catalog tables.
Gemini Omni Flash in public preview
Released gemini-omni-flash-preview, a high-performance multimodal model designed for high-speed video generation and conversational video editing. Using the Interactions API, you can generate 3--10 second videos at 720p from text descriptions or animate still images, and then conversationally edit and refine the outputs.
Gemini 3.1 Flash Lite Image (Nano Banana 2 Lite) is generally available
gemini-3.1-flash-lite-image is a multimodal model optimized for ultra-low latency, cost-effective image generation and editing, now GA in the Gemini API.
Claude Sonnet 5 launches with a 1M token context window at $2 / $10 per MTok
Claude Sonnet 5 supports a 1M token context, 128k max output and the same tools as Sonnet 4.6 except Priority Tier. Migrating teams should note that adaptive thinking is on by default and manual extended thinking changes.
More from Microsoft FoundryEvery Microsoft Foundry release →
[Launched] Public Preview: Document PII playground sample in Microsoft Foundry NextGen
Document-based PII detection in Azure AI Language now ships an initial playground experience in the Microsoft Foundry NextGen portal. The playground loads a prepared sample document, runs it through the document-based PII feature, and shows the redacted o
[Launched] Generally Available: Document PII NextGen Playground in Azure AI Language
Azure AI Language now offers a refreshed Document PII playground in general availability, giving customers a faster way to evaluate personally identifiable information detection on documents. The playground ships with curated sample inputs and outputs, so
Announcing: Multi-agent orchestration patterns including Magentic in Microsoft Agent Framework
Microsoft Agent Framework reaches general availability for multi-agent orchestration patterns, including the dynamic Magentic pattern that delivered 38 percent on GAIA and competitive results on AssistantBench and WebArena in published benchmarks. The GA
Announcing: Agent Harness in Microsoft Agent Framework
Microsoft Agent Framework reaches general availability for Agent Harness, the production runtime for agents built with the framework. Customers running pilots on the preview harness can now move the same agent code into production with backwards compatibi
Announcing: Multi-agent orchestration SDK with the Microsoft Foundry Agent Framework in C# and Python
Microsoft Foundry is now introducing Agent Framework, a unified multi-agent orchestration SDK for C# and Python. This release consolidates patterns previously spread across AutoGen and Semantic Kernel, giving developers a consistent set of abstractions fo
Announcing: Tracing for Microsoft Agent Framework (Python and .NET)
In public preview, Microsoft Foundry now offers tracing for the Microsoft Agent Framework in both Python and .NET, helping developers observe how agents reason, call tools, and interact with models during execution. This release is part of the broader hos
Announcing: CodeAct pattern and Hyperlight containers in Microsoft Agent Framework
Microsoft Agent Framework adds the CodeAct pattern in public preview, which lets agents collapse a multi-step plan into a single executable code block rather than chaining one tool call at a time. In representative workloads, CodeAct cut end-to-end latenc
Announcing: Episodic procedural memory in Microsoft Agent Framework
Microsoft Agent Framework adds episodic procedural memory in public preview, a built-in store that lets agents remember successful task patterns and replay them on similar future requests. The framework records action sequences, tool selections, and inter