Microsoft released version 1.0 of the microsoft.foundry extension bundle for Azure Developer CLI. It scaffolds hosted agents, keeps agent and deployment configuration in azure.yaml, supports local inspection and testing, and carries the same configuration into CI and deployment.
Why it matters: Agent configuration can now be reviewed with application code and promoted through repeatable pipelines instead of being recreated manually in a portal, improving change control and rollback discipline for hosted-agent deployments.
Microsoft Agent Framework for Go 0.2.0 added common embedding-generation and typed vector-store APIs. Developers can start with in-memory storage and move to Azure AI Search or Cosmos DB while preserving the retrieval interface used to supply agent context and memory.
Why it matters: A shared Go retrieval abstraction can reduce backend-specific agent code and make migration easier, but abstractions can hide schema, dimensionality, consistency, security and billing differences that determine production behavior.
Microsoft added Microsoft-Decision-1 to the Foundry catalog in public preview. Built by post-training Qwen3.5-9B, it takes predefined choices and returns bounded answers with confidence signals for routing, classification, evaluation and workflow control.
Why it matters: Typed choices can reduce parsing failures and cost in high-volume control steps where free-form generation is unnecessary. The confidence output is only useful if it is calibrated on the actual decision set and paired with abstention or human review.
Anthropic released Claude Haiku 5.5 across its API, apps and major cloud platforms. The small model adds adjustable effort and targets summaries, classification, browser use and coding subagents at $0.10 per million input tokens and $0.50 per million output tokens for prompts up to 100,000 tokens.
Why it matters: The pricing makes delegated model calls materially cheaper, so teams can reserve stronger models for hard reasoning while moving repetitive classification, compaction and tool work to a smaller tier.
Cohere released Embed 5 Pro and Fast for text, image and fused retrieval across more than 100 languages and contexts up to 128K tokens. Both models share an embedding space, allowing teams to index with Pro and query with Fast without rebuilding the index, and support Matryoshka dimensions plus float, int8 and binary outputs.
Why it matters: A compatible quality and latency tier can reduce query-path cost without forcing a second corpus index. The multimodal and compressed-output options also make storage, retrieval quality and private-serving requirements explicit engineering choices rather…
Microsoft's Visual Studio 2026 stable-channel update enables Bring Your Own Model by default across Community, Professional and Enterprise. Developers can connect Microsoft Foundry, OpenAI, Anthropic, Ollama or supported custom endpoints to a new GitHub Copilot SDK-powered Agent preview; the older BYOM path is no longer supported and existing Ollama models must be added again.
Why it matters: The update changes both the model boundary and the execution harness inside a mainstream IDE. Teams can use approved or local endpoints, but existing BYOM configurations and agent behavior can change after the stable-channel upgrade.
Microsoft Research released RetroChimera code and weights under an MIT license alongside a Nature paper. The system ensembles a generative reaction model with a template-based model and learns to rerank their proposals for single-step retrosynthesis.
Why it matters: Retrosynthesis is a high-value scientific workflow where rare reaction recall and out-of-distribution robustness matter more than fluent output. An inspectable model and checkpoints let chemistry teams test those claims on their own compounds instead of…
Claude Code 2.1.278 makes its server-side auto-mode classifier the default for Claude API and Enterprise users and for Bedrock, Vertex, Foundry and gateway sessions. Anthropic says server-side classification does not charge classifier overhead; affected gateway deployments can opt out, and the CLI now reports whether the server path is active.
Why it matters: Moving a permission classifier changes both billing and a control-path dependency. A proxy or gateway that cannot preserve the required request and response fields may fall back to the billed local path, so teams need evidence that the advertised mode is…
Anthropic released Apache-2.0 reference implementations for shopping and merchant agents across retail, travel, telecom and entertainment. The shopping agent searches, compares and prepares carts while leaving payment to the retailer. The merchant agent reads sales and inventory and drafts pricing, promotion and campaign changes, with writes staged for human approval. The repository supports the Claude API and major cloud channels.
Why it matters: Teams can inspect and run an end-to-end agent pattern that separates research, recommendation, payment and merchant-side writes instead of designing those authority boundaries from scratch.
Cohere made Parse v5.0 generally available through its API, Model Vault, Microsoft Foundry, and AWS SageMaker. The vision parser turns supported documents and images into reading-order Markdown with table and image handling across nine languages; the API launch price is $1.50 per 1,000 pages.
Why it matters: Document parsing is the evidence boundary for retrieval and document agents. A fast or cheap parser is useful only if it preserves tables, reading order, formatting, access controls, page provenance, and failure signals well enough for the downstream decision.
Microsoft opened MAI-Thinking-1 in Foundry public preview. The sparse mixture-of-experts model has 35 billion active and roughly one trillion total parameters, a 256,000-token context window, function calling, developer instructions, and Chat Completions compatibility.
Why it matters: Public access turns an earlier partner-only model announcement into a testable enterprise reasoning endpoint with a smaller active inference footprint.
GitHub confirmed that Models is fully retired. Its playground, model catalog, inference API and bring-your-own-key endpoints are unavailable to every customer, including existing users with active workloads.
Why it matters: Any integration still calling GitHub Models is now broken, and Copilot is not a drop-in replacement for a general inference API.
GitHub added BYOK support for OpenAI, Azure, Microsoft Foundry, Anthropic, LM Studio, Ollama, and OpenAI-compatible endpoints, storing local secrets in the operating system keychain.
Why it matters: Developer-agent interfaces are separating from model procurement, enabling private, local, and enterprise-controlled inference behind one workflow.
At Build, Microsoft introduced or expanded seven MAI models, including the 35B-active-parameter MAI-Thinking-1 and new image, transcription, voice, and low-latency coding models. Several media models reached Foundry while the reasoning model stayed with select early partners.
Why it matters: Microsoft is moving from being primarily a distributor of partner models toward owning a vertically integrated portfolio for distinct production workloads.
Anthropic released Claude Opus 4.7 across Claude, its API, Amazon Bedrock, Google Vertex AI, and Microsoft Foundry at 5 dollars per million input tokens and 25 dollars per million output tokens. The update targets long-running coding and professional work and adds higher-resolution image input, an xhigh effort level, task budgets in beta, and an ultrareview mode.
Why it matters: A leading model vendor is exposing explicit effort and budget controls for long-running agents, making inference allocation a first-class product decision.
Microsoft made its next-generation Foundry Agent Service and Foundry Control Plane observability generally available while putting Voice Live integration into public preview. It also said Azure had powered on Vera Rubin NVL72 in its labs, planned rollout over the following months, and released a public Physical AI Toolchain repository.
Why it matters: Cloud platforms are combining governed agent runtimes with next-generation inference infrastructure and simulation-to-operations tooling for physical systems.
Anthropic released Claude Sonnet 4.6 with a one-million-token context window in beta and made it the default model for Free and Pro users as well as Cowork. It was priced at $3 per million input tokens and $15 per million output tokens across the API and major clouds.
Why it matters: A strong workhorse model can create more product value than a premium flagship if it handles routine coding, computer use and document work with fewer escalations. Default placement also exposes new capabilities to a much larger real-world workload mix.
Anthropic released Claude Opus 4.6 with a one-million-token context window in beta, adaptive thinking and effort controls, context compaction, agent teams and improved document tooling. It was available in Claude, the API and major cloud platforms at $5 per million input tokens and $25 per million output tokens.
Why it matters: Long context is becoming useful when paired with compaction and sustained tool use, not when treated as a document dump. Effort controls also let teams spend more inference only on tasks where additional reasoning changes the outcome.
Microsoft released Fara-7B under an MIT license, a seven-billion-parameter vision-language model that controls webpages through screenshots and predicted mouse and keyboard actions.
Why it matters: A locally deployable computer-use model offers lower latency and stronger data locality than browser agents requiring repeated frontier API calls.