Hugging Face Models on Foundry Managed Compute

At Microsoft Build 2026, we announced Foundry Managed Compute and Hugging Face models on Foundry — a curated catalog of open-weight models from the Hugging Face ecosystem, refreshed weekly, deployable in one click onto Foundry Managed Compute. Weights are pre-staged in Azure, runtimes are built and scanned by Microsoft, and every model in the Collection ships with the same enterprise security, governance, observability, and billing that applies to every other model on Foundry.

在 2026 年微软 Build 大会上，我们发布了 Foundry Managed Compute 以及 Foundry 上的 Hugging Face 模型——这是一个精选的 Hugging Face 生态系统开源权重模型目录，每周更新，并支持一键部署到 Foundry Managed Compute 上。模型权重已预先部署在 Azure 中，运行时环境由微软构建并经过扫描，集合中的每一个模型都具备与 Foundry 上其他所有模型相同的企业级安全性、治理、可观测性和计费标准。

The Platform: Microsoft Foundry and Managed Compute

平台：Microsoft Foundry 与 Managed Compute

Microsoft Foundry is a platform for building and operating agentic AI applications. Foundry starts with the widest model selection on any cloud — models from Microsoft, OpenAI, Anthropic, Meta, Mistral, DeepSeek, Hugging Face, and others, spanning frontier, open-source, and custom weights — all accessible through a single endpoint and a single set of SDKs in Python, C#, JavaScript, and Java.

Microsoft Foundry 是一个用于构建和运行智能体（Agentic）AI 应用的平台。Foundry 提供了云端最广泛的模型选择——涵盖来自微软、OpenAI、Anthropic、Meta、Mistral、DeepSeek、Hugging Face 等厂商的前沿模型、开源模型及自定义权重模型。所有这些模型均可通过单一端点以及 Python、C#、JavaScript 和 Java 的统一 SDK 集进行访问。

On top of those models sits the Foundry Agent Service: multi-agent orchestration with built-in memory, knowledge grounding through Foundry IQ, and a catalog of connectable tools via agentic protocols, so agents can work with enterprise data. Once agents are running, Foundry provides end-to-end tracing, real-time monitoring, continuous evaluations, and a prompt optimizer that improves agent behavior based on eval results — observability and quality loops that are part of the platform.

在这些模型之上是 Foundry Agent Service：它提供具备内置记忆功能的多智能体编排、通过 Foundry IQ 实现的知识基础（Knowledge Grounding），以及通过智能体协议连接的工具目录，使智能体能够处理企业数据。当智能体运行后，Foundry 提供端到端追踪、实时监控、持续评估以及基于评估结果优化智能体行为的提示词优化器——这些可观测性和质量闭环都是该平台的一部分。

Alongside that, developers get access to:

Content safety filters
Task-adherence guardrails
An AI Red Teaming Agent for adversarial testing
Unified RBAC
Private networking
Azure Policy integration directly within the platform

此外，开发者还可以获得以下功能：

内容安全过滤器
任务遵循护栏（Task-adherence guardrails）
用于对抗性测试的 AI 红队智能体
统一的基于角色的访问控制（RBAC）
私有网络
直接集成在平台内的 Azure Policy

Alongside pay-per-token (lowest-friction path to get started) and provisioned throughput (predictable, high-performance production workloads on frontier models), Foundry Managed Compute is the third deployment option in Foundry: a managed GPU platform-as-a-service for open-source and custom models. You deploy a model instance described by the things that matter to your workload — parameter count, context length, and whether you want to optimize for latency or throughput — and Foundry handles the GPU topology underneath, whether the instance lands on one accelerator or several, so you think and plan in model terms.

除了按 Token 计费（最简单的入门路径）和预置吞吐量（针对前沿模型的可预测、高性能生产工作负载）之外，Foundry Managed Compute 是 Foundry 中的第三种部署选项：这是一个面向开源和自定义模型的托管 GPU 平台即服务（PaaS）。你只需根据工作负载的关键指标（如参数量、上下文长度，以及是优化延迟还是吞吐量）来部署模型实例，Foundry 会在底层处理 GPU 拓扑结构，无论实例是运行在一个还是多个加速器上，你只需从模型层面进行思考和规划。

Microsoft takes care of the machine: container updates, runtime upgrades, and security patches happen automatically on the supported runtimes — vLLM, SGLang, TensorRT-LLM, NIM, TEI, llama.cpp — without redeploying your model, while model configuration, deployment behavior, and routing stay with you. That consistency carries through the developer surface — pay-per-token, provisioned throughput, and Managed Compute share:

A single endpoint
The same SDKs
The same authentication
The same observability
A single bill

微软负责机器维护：容器更新、运行时升级和安全补丁会在支持的运行时（vLLM、SGLang、TensorRT-LLM、NIM、TEI、llama.cpp）上自动完成，无需重新部署模型，而模型配置、部署行为和路由控制权则保留在用户手中。这种一致性贯穿于开发者界面——按 Token 计费、预置吞吐量和 Managed Compute 共享：

单一端点
相同的 SDK
相同的身份验证
相同的可观测性
统一账单

Open-source models integrate with Foundry Agents the same way frontier models do, so you can mix model types in a single agent without a separate integration path. Managed Compute offers:

Global deployments — broadest capacity and best pricing
Data Zone deployments — residency and sovereignty
Same code, same workflow. Quota is aligned to accelerator families, so a plan built on the H100 family today carries forward as new hardware generations come online.

开源模型与 Foundry Agents 的集成方式与前沿模型相同，因此你可以在单个智能体中混合使用不同类型的模型，而无需额外的集成路径。Managed Compute 提供：

全球部署——最广泛的容量和最优的价格
数据区域部署——满足数据驻留和主权要求
相同的代码，相同的工作流。配额与加速器系列对齐，因此今天基于 H100 系列构建的计划，在未来新一代硬件上线时依然适用。

Why Hugging Face

为什么选择 Hugging Face

Hugging Face is the public square of open AI: 15 million builders, 400,000 organizations, and over 3 million open models published, with new frontier capabilities — agentic coding, video segmentation, speech, embeddings — landing weekly. It’s the GitHub of open models, where the community publishes weights, writes model cards, compares evaluations, and pulls models for experimentation.

Hugging Face 是开放 AI 的公共广场：拥有 1500 万开发者、40 万家组织，发布了超过 300 万个开源模型，每周都有新的前沿能力（如智能体编码、视频分割、语音、嵌入等）落地。它是开源模型的 GitHub，社区在这里发布权重、编写模型卡、对比评估结果并拉取模型进行实验。

Open models have closed the gap with proprietary models on benchmark after benchmark, and they unlock things proprietary endpoints can’t:

State-of-the-art is now open. Leading open-weight models are competitive with the top closed frontier models on the most widely used benchmarks.
Deep customization. Full weights make it possible to fine-tune, distill, quantize, and adapt with LoRA — tailoring models to your domain, your data, and your latency and cost targets.
Your model, your hosting. Weights run in your tenant on infrastructure you control, behind your inference endpoint, with your identity and network boundaries.
Cost shaping. Pay for accelerators by the hour, scale to zero when idle, and right-size GPUs to the specific model — useful for steady, high-volume, or latency-sensitive workloads where per-token pricing is harder to predict.
Version control. Pin a specific model version, evaluate it, deploy it, and move forward or roll back on your own release cadence.

开源模型在各项基准测试中已缩小了与专有模型的差距，并解锁了专有端点无法实现的功能：

最先进的技术现已开源。 领先的开源权重模型在最广泛使用的基准测试中，已能与顶尖的闭源前沿模型相媲美。
深度定制。 完整的权重使得微调、蒸馏、量化和使用 LoRA 进行适配成为可能，从而根据你的领域、数据以及延迟和成本目标来定制模型。
你的模型，你的托管。 权重运行在你控制的基础设施上的租户中，位于你的推理端点之后，并具备你的身份和网络边界。
成本塑造。 按小时为加速器付费，闲置时缩减至零，并根据特定模型调整 GPU 大小——这对于按 Token 计费难以预测的稳定、高容量或延迟敏感型工作负载非常有用。
版本控制。 锁定特定模型版本，进行评估、部署，并按照你自己的发布节奏进行升级或回滚。

The catch has always been the operational layer: discovery, license review, security screening, runtime selection, GPU sizing, image building, CVE patching, and standing the model up behind an enterprise-grade endpoint. Hugging Face, by itself, is not an enterprise serving platform. Hugging Face models on Foundry is that operational layer, run by Microsoft.

一直以来的难点在于运营层：发现模型、许可审查、安全筛选、运行时选择、GPU 大小调整、镜像构建、CVE 补丁修复，以及在企业级端点后部署模型。Hugging Face 本身并非企业级服务平台。而 Foundry 上的 Hugging Face 模型正是由微软运营的这一运营层。

Hugging Face Models on Foundry

Foundry 上的 Hugging Face 模型

The Hugging Face Collection brings a curated subset of models directly into the Foundry Model Catalog:

Refreshed weekly — trending models from the Hugging Face ecosystem are added continuously as the community publishes them.
Every modality — text, vision, audio, and multimodal: LLMs and VLMs for chat and agents, ASR and speech translation, embeddings, segmentation, image generation.
Safetensors only, no untrusted code — every model in the Collection is security-screened and ships in the SafeTensors weight format, with no trust_remote_code execution paths unless rigorously reviewed.
The right runtime for the model — vLLM and SGLang for LLMs, TensorRT-LLM and NIM where applicable, TEI for embeddings, llama.cpp for CPU — Foundry picks the engine that matches the model.

Hugging Face 集合将精选的模型子集直接引入 Foundry 模型目录：

每周更新——随着社区发布，Hugging Face 生态系统中的热门模型会持续添加。
全模态支持——文本、视觉、音频和多模态：用于聊天和智能体的 LLM 和 VLM、ASR 和语音翻译、嵌入、分割、图像生成。
仅限 Safetensors，无不可信代码——集合中的每个模型都经过安全筛选，并以 SafeTensors 权重格式发布，除非经过严格审查，否则不包含 trust_remote_code 执行路径。
匹配模型的正确运行时——LLM 使用 vLLM 和 SGLang，适用场景使用 TensorRT-LLM 和 NIM，嵌入使用 TEI，CPU 使用 llama.cpp——Foundry 会自动选择与模型匹配的引擎。