|
| 1 | +# MSFT Foundry - Overview |
| 2 | + |
| 3 | +Costa Rica |
| 4 | + |
| 5 | +[](https://github.com/) |
| 6 | +[brown9804](https://github.com/brown9804) |
| 7 | + |
| 8 | +Last updated: 2026-01-05 |
| 9 | + |
| 10 | +------------------------------------------ |
| 11 | + |
| 12 | +> Microsoft Foundry is not just a rebrand, it’s a major evolution. `It shifts from model-centric experimentation to agent-centric enterprise AI, |
| 13 | +> with unified governance, observability, and developer tooling.` If you used Classic Foundry, expect a more streamlined portal, |
| 14 | +> stronger compliance features, and a deeper focus on agents. |
| 15 | +
|
| 16 | +<details> |
| 17 | +<summary><b>List of References</b> (Click to expand)</summary> |
| 18 | + |
| 19 | +- [What is Microsoft Foundry?](https://learn.microsoft.com/en-us/azure/ai-foundry/what-is-foundry?view=foundry-classic&utm_source=copilot.com) |
| 20 | +- [Microsoft Foundry architecture](https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/architecture?view=foundry-classic&utm_source=copilot.com) |
| 21 | +- [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 | +- [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 | + |
| 24 | +</details> |
| 25 | + |
| 26 | +| MSFT Foundry (New) | AI Foundry (Classic) | |
| 27 | +| --- | --- | |
| 28 | +| <img width="1901" height="982" alt="image" src="https://github.com/user-attachments/assets/d7195178-8d6d-4681-90f2-8fd273ed95f7" /> | <img width="1893" height="980" alt="image" src="https://github.com/user-attachments/assets/89769a23-376b-4f7e-81aa-e9ea73b0258d" /> | |
| 29 | + |
| 30 | +## Classic vs. New Foundry |
| 31 | + |
| 32 | +> - Unified AI Platform: Combines `infrastructure, models, agents, and tools under one management layer` |
| 33 | +> - Enterprise Ready: `Built-in monitoring, tracing, evaluations, and compliance features` |
| 34 | +> - Agent Centric: Designed around `AI agents and workflows rather than just models` |
| 35 | +> - Developer-Friendly: Simplifies `project setup, resource management, and SDK/API usage` |
| 36 | +
|
| 37 | +`Classic = LLM sandbox` |
| 38 | +`New = Agent platform with enterprise governance` |
| 39 | + |
| 40 | +| Aspect | Foundry (Classic) | Foundry (New) | |
| 41 | +|--------|------------------|---------------| |
| 42 | +| **Focus** | `LLM deployment & experimentation` → Classic Foundry was designed around large language models. Developers could explore different LLMs, fine‑tune them with custom data, and embed them into applications. It was essentially a sandbox for working with generative AI models, not a full enterprise agent platform. | `Agent-first workflows, governance, observability` → The new Foundry shifts from just hosting LLMs to managing **AI agents** that use LLMs as components. It emphasizes workflows, compliance, monitoring, and enterprise‑scale deployment. Models are still there, but they’re part of a larger agent ecosystem. | |
| 43 | +| **Portal Experience** | `Fragmented resource creation` → Users had to separately configure model endpoints, storage, and monitoring. | `Streamlined project creation & unified portal` → One project setup provisions everything needed for agents, models, and observability. | |
| 44 | +| **Governance** | `Limited enterprise controls` → Basic access and monitoring, but not deep compliance. | `Full enterprise setup: tracing, monitoring, evaluations` → Governance is built in, with dashboards for performance, compliance, and policy enforcement. | |
| 45 | +| **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 | +| **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 | +<!-- START BADGE --> |
| 49 | +<div align="center"> |
| 50 | + <img src="https://img.shields.io/badge/Total%20views-1497-limegreen" alt="Total views"> |
| 51 | + <p>Refresh Date: 2026-01-05</p> |
| 52 | +</div> |
| 53 | +<!-- END BADGE --> |
0 commit comments