You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
-[What is the Microsoft Foundry Control Plane?](https://learn.microsoft.com/en-us/azure/ai-foundry/control-plane/overview?view=foundry&utm_source=copilot.com)
22
22
-[Comparing Content Understanding in Microsoft Foundry vs Content Understanding Studio](https://learn.microsoft.com/en-us/azure/ai-services/content-understanding/foundry-vs-content-understanding-studio?utm_source=copilot.com)
23
+
-[Foundry Agent Service at Ignite 2025: Simple to Build. Powerful to Deploy. Trusted to Operate.](https://techcommunity.microsoft.com/blog/azure-ai-foundry-blog/foundry-agent-service-at-ignite-2025-simple-to-build-powerful-to-deploy-trusted-/4469788)
23
24
24
25
</details>
25
26
@@ -34,7 +35,7 @@ Last updated: 2026-01-05
34
35
> - Agent Centric: Designed around `AI agents and workflows rather than just models`
|**Integration**|`Separate tools for models, apps, infra` → Developers had to stitch together SDKs and services. |`Unified under one platform (agents + models + tools)` → A cohesive environment where agents, models, and data connectors work seamlessly. |
46
47
|**Evolution**|`Azure AI Foundry branding` → Focused on LLM experimentation within Azure. |`Microsoft Foundry rebrand (Ignite 2025)` → Signals the strategic pivot to enterprise AI agents and unified workflows. |
47
48
49
+
## Key New Features (2025–2026)
50
+
51
+
> `Microsoft Foundry is no longer just a model playground, it’s a unified agent platform that lets developers and enterprises`:
52
+
53
+
- Build multi-agent systems
54
+
- Orchestrate models across providers
55
+
- Govern and monitor AI behavior
56
+
- Deploy agents into real-world environments like Teams, web apps, and internal tools
57
+
58
+
| Feature | Description | Key Benefits |
59
+
|---------|-------------|--------------|
60
+
|[Hosted Agents](https://learn.microsoft.com/en-us/azure/ai-foundry/agents/concepts/hosted-agents?view=foundry&tabs=cli)| Build, test, and deploy agents without managing infrastructure. No need for Kubernetes or containers. Supports multi-agent orchestration with shared memory and tools. | Simplifies deployment, reduces infra overhead, enables complex agent ecosystems for enterprise workflows and internal copilots. |
61
+
|[Model Router](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/concepts/model-router?view=foundry-classic#supported-underlying-models)| Routes requests across multiple models (OpenAI, Claude, Gemini, etc.). Automatically selects the best model for the task. Supports fallback and load balancing. | Optimizes performance and cost, ensures reliability, allows seamless switching between models without code changes. |
62
+
|[BYO Model Gateway](https://learn.microsoft.com/en-us/azure/ai-foundry/agents/how-to/ai-gateway?view=foundry)| Bring your own model (BYOM) from Hugging Face, Azure ML, or custom endpoints. Integrates with Foundry’s governance and observability stack. | Flexibility to use custom or proprietary models, hybrid cloud/on-prem strategies, unified monitoring and compliance. |
63
+
|[One-Click Deployment to Microsoft 365](https://learn.microsoft.com/en-us/azure/ai-foundry/agents/how-to/publish-copilot?view=foundry)| Deploy agents directly into Teams, Outlook, SharePoint, and other M365 apps. Includes low-code/no-code templates for internal copilots. Supports user identity, permissions, and organizational data access. | Rapid enterprise adoption, seamless integration with daily productivity tools, secure identity-aware deployments. |
64
+
| [Built-in Observability](https://techcommunity.microsoft.com/blog/azure-ai-foundry-blog/observability-in-foundry-control-plane-empowering-developers-to-evaluate-and-opt/4471107) | Native tracing, telemetry, and evaluation dashboards. Human-in-the-loop feedback tools. Agent memory inspection and optimization tracking. | Improves reliability, enables debugging and optimization, ensures agents meet performance and compliance standards. Useful refs: [Exercise 04: Observe model utilization in Microsoft Foundry](https://microsoft.github.io/TechWorkshop-L300-AI-Apps-and-agents/docs/04_observability_ai_foundry/04_observability_ai_foundry.html), [Agent tracing overview (preview)](https://learn.microsoft.com/en-us/azure/ai-foundry/observability/concepts/trace-agent-concept?view=foundry), [Observability in generative AI](https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/observability?view=foundry-classic), [Continuously evaluate your AI agents (preview)](https://learn.microsoft.com/en-us/azure/ai-foundry/how-to/continuous-evaluation-agents?view=foundry-classic) |
65
+
|[Enterprise Governance](https://learn.microsoft.com/en-us/azure/ai-foundry/control-plane/overview?view=foundry)| Identity management via Microsoft Entra ID. Policy enforcement, audit logging, and compliance controls. Role-based access and agent sandboxing. | Enterprise-grade security, compliance readiness, controlled access, and safe scaling across organizations. |
From [Bring your own AI gateway to Azure AI Agent Service (preview)](https://learn.microsoft.com/en-us/azure/ai-foundry/agents/how-to/ai-gateway?view=foundry)
0 commit comments