Showing tag results for evaluations

9 min read · June 3, 2026 · Sebastian Kohlmeier    Shipping an AI agent is the easy part. Keeping it accurate, safe, and accountable in production is where teams get stuck. Agents are non-deterministic. Their behavior shifts as models update, tools change, and traffic patterns evolve and most of that drift happens silently, long after the dem...

Learn a practical model lifecycle for Microsoft Foundry: select the right model, evaluate quality, optimize cost, operate safely, and improve as production needs change.

Learn how Microsoft helps developers build trustworthy AI agents with open evaluations, portable runtime controls, production observability, and security workflows that work across frameworks.

Microsoft Build 2026 brings a major set of Microsoft Foundry updates for developers building agents: hosted runtimes, Toolboxes, memory, Voice Live, Foundry IQ, new models, managed compute, and trust, evaluation, and observability tools.

May ships trace-based evaluation for any agent on any cloud, Grok 4.3 and DeepSeek V4 in the model catalog, GPT-5 Reinforcement Fine-Tuning at gated GA, three Microsoft Research on-device agent models, Managed VNET at GA, project-level cost attribution, Content Understanding improvements reaching GA, Foundry Local 1.1 and 1.2 with live audio and vi...

Walk through running quality, cost, and latency evaluations for the Foundry model router using an open-source GitHub repo designed for router-aware eval pipelines.

April brings a wave of model arrivals — GPT-5.5, GPT-image-2, Microsoft first-party MAI models for image, voice, and transcription, Gemma 4, and Claude Opus 4.7 — alongside Foundry Local GA, Microsoft Agent Framework 1.0 GA, the Microsoft Foundry Toolkit for VS Code GA, batch evaluations for third-party agents, new tracing and monitoring capabiliti...

March ships Foundry Agent Service GA with private networking, GPT-5.4 and GPT-5.4 Mini, Priority Processing, Phi-4 Reasoning Vision, SDK 2.0 GA across Python, JS/TS, Java, and .NET, Fireworks AI and NVIDIA Nemotron models, and third-party guardrails from Palo Alto and Zenity.

The next-gen Foundry Agent Service is generally available today with end-to-end private networking, Voice Live integration, expanded MCP authentication, GA evaluations with continuous monitoring, and hosted agent deployments in six new Azure regions.

GPT‑5‑Codex GA, Sora video‑to‑video preview, Grok 4 Fast, Browser Automation & Key Vault previews, new Search knowledge sources, Voice Live GA.