Azure AI Foundry Blog
Your source for learning and building with our models, agents, and tools.
Latest posts
A Developer’s Guide to Fine-Tuning GPT-4o for Image Classification on Azure AI Foundry
Whether you’re a machine learning practitioner, app developer, or just curious about the latest in AI, this guide shows how you can quickly boost image classification accuracy using cutting-edge Vision-Language Models (VLM) on Azure—no deep learning expertise required. In this walkthrough, you’ll see how to fine-tune GPT-4o on Azure OpenAI for image classification using the Stanford Dogs dataset. Illustrations of some of dogs' breeds images presented on the dataset. We’ll use the Vision Fine-Tuning API and compare the results to a lightweight CNN baseline, so you can see the impact of modern Vision-Languag...
The Developer’s Guide to Smarter Fine-tuning: Unlock custom AI for every business challenge
Azure AI Foundry makes fine-tuning smarter, faster, and more accessible than ever. Whether you’re building agents that reason, tools that adapt, or workflows that scale, this is your launchpad for customizing models to solve real business challenges. Dive in to discover best practices, hands-on resources, and the latest innovations so you can build, test, and deploy specialized AI with confidence. What is Fine-tuning? Fine-tuning refers to customizing a pre-trained LLM with additional training on a specific task or new dataset for enhanced performance, new skills, or improved accuracy. So instead of building ...
What’s new in Azure AI Foundry | September 2025
GPT‑5‑Codex GA, Sora video‑to‑video preview, Grok 4 Fast, Browser Automation & Key Vault previews, new Search knowledge sources, Voice Live GA.
Introducing Microsoft Agent Framework: The Open-Source Engine for Agentic AI Apps
Why agents need a new foundation Over the last year, developers have been experimenting with AI agents in every imaginable form. Agents are not just chatbots or copilots — they are autonomous software components that can reason about goals, call tools and APIs, collaborate with other agents, and adapt dynamically. Whether it’s a retrieval agent for research, a coding agent embedded in a dev workflow, or a compliance agent ensuring policy enforcement, agents are becoming the next layer of application logic. Yet despite the excitement, the path from prototype to production has been fraught with obstacles. Many ...
Foundry Local Meets More Silicon
Foundry Local is a high-performance local AI runtime stack that brings Azure AI Foundry’s power to client devices. Foundry Local lets you build and ship cross-platform AI apps that run models with acceleration on a wide range of hardware. The Evolution of AI Acceleration On-device AI has progressed rapidly. Early workloads ran on CPUs, but performance and power limits made real-time inference difficult. GPUs improved things with parallelism, faster inference. The latest breakthrough is NPUs (Neural Processing Units), designed specifically for neural networks. NPUs deliver far greater efficiency and throughput...
AI-Assisted Development powered by Local Models
Developers have long struggled with concerns about data privacy, restrictive cloud dependencies, and limited control over their AI tools—especially when working on sensitive projects or within regulated industries. Now, with the excitement surrounding local models, you can harness AI directly on your device, ensuring your data stays private and your workflow remains fully customizable. The big breakthrough: Foundry Local empowers you to use local AI models seamlessly with GitHub Copilot in Visual Studio Code, giving you privacy, flexibility, and offline capabilities without compromise. Foundry Local is a cutti...
Ground Your Agents Faster with Native Azure AI Search Indexing in Foundry
Instantly create a production-ready Azure AI Search vector index directly inside Azure AI Foundry when grounding your agent—no prior search setup required.