Azure OpenAI in Foundry Models

by Microsoft@microsoftRedmond, Washington, USANo ratings yet

Azure-hosted access to OpenAI models for building apps and agents

Tool type: Hosted OpenAI model API on Azure
Made for: Developers and enterprise teams
Runs on: Azure cloud API endpoints; Models are called through Azure-hosted deployments; the Microsoft Foundry portal is used to deploy and manage them.
Last checked: 28 Oct 2026; 2 product sources

What it does

Azure OpenAI in Foundry Models gives organizations API access to OpenAI's text, reasoning, image, audio, and video models hosted on Microsoft Azure. Developers send prompts, audio, images, or documents to a deployed model and receive text, speech, images, or video in return, with options for fine-tuning and connecting models to agents.

Teams typically use it to build customer support tools, summarize conversations, generate marketing or product content, analyze proprietary data, and automate routine workflows. Its distinguishing points on the page are Azure-level data privacy controls, regional and global deployment choices, built-in content safety tooling, and integration with other Azure services such as Azure AI Search and Foundry Agent Service.

This suits developers and enterprises already working in the Azure ecosystem who need compliance controls and predictable capacity. Access requires an Azure account and a deployed model; pricing is pay-as-you-go by token or provisioned throughput. Many model prices are not visible in the captured text, and the service is described as having limited access for some use cases.

What it can do

Text generation and reasoning with OpenAI GPT model series
Multimodal input including images and documents
Real-time and speech audio models for voice applications
Image generation and editing models
Fine-tuning to customize models for specific business needs
Built-in content moderation and responsible AI tooling

Works with

Azure AI Search
Foundry Agent Service
Azure Cosmos DB
Microsoft Fabric

Example workflows

Summarize customer support calls: Input: call transcript text. Steps: send the transcript to a deployed GPT model with a summarization prompt; receive a structured summary with issue and resolution. Output: concise call summary for the support record.
Answer questions over internal documents: Input: a user question and documents indexed in Azure AI Search. Steps: retrieve relevant passages, send them with the question to a GPT model, return the answer. Output: a grounded answer that references source documents.

Before you choose

Pricing varies by model, deployment type, and context length, and many current rates are not shown in the captured page; check the pricing page or calculator for current values.
Some use cases are subject to limited access and approval criteria covering responsible use and capacity.
Pricing estimates on the page are not binding quotes and may vary by Microsoft agreement and exchange rates.

Listed under

Get started

Create or sign in to an Azure account at the Azure sign-in page.
Open Microsoft Foundry and deploy a chosen OpenAI model to a region or deployment type (Standard, Provisioned, or Batch).
Call the deployed model from your application using the endpoint and credentials provided in the deployment.

Create an Azure account and deploy a model

Plans and prices

Prices in USD

Standard is pay-as-you-go per input and output token; Provisioned uses PTUs with optional monthly or annual reservations; Batch offers a 50% discount on Global Standard pricing for completions within 24 hours. Exact rates for most models were not captured, so check the pricing page or calculator.

Standard (pay-as-you-go)Usage-basedRates vary by model; see the pricing page or calculator for current rates.

Pay per input and output token
Global, Data Zone, and Regional deployment options

Provisioned (PTU)Usage-basedPricing is per PTU; check the pricing calculator for current values.

Reserved throughput with predictable cost
Monthly and annual reservations available

Source: azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service, checked 28 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 28 Oct 2026. Each source supports part of the overview; availability and prices can change.

Azure OpenAI in Foundry Models | Microsoft Azure ↗azure.microsoft.com
Azure OpenAI Service - Pricing | Microsoft Azure ↗azure.microsoft.com

Ratings and reviews

No ratings for Azure OpenAI in Foundry Models yet.

Be the first to rate it

What’s changed

Free account

See what’s changed at Azure OpenAI in Foundry Models

Launches, price moves and pivots, each with a date and a source. Free with an account.

Signals

Free account

See Azure OpenAI in Foundry Models’s signals

Launch traction, hiring and release activity, tracked over time. Free with an account.

Data and security

Free account

See how Azure OpenAI in Foundry Models handles your data

Training on your files, compliance, hosting and the models behind it. Free with an account.

Compare with similar tools

Free account

Compare Azure OpenAI in Foundry Models side by side

Prices, platforms and features against similar tools, with sources. Free with an account.

Where this comes from

How we research

Company website: azure.microsoft.com: product, pricing and policy pages
YC directory mirror: Batch and company name
Web search: Confirms the official site
Social accounts: Accounts the company links, and its posts on X
Human review: Checked by a person
knowledge_base

Last checked 28 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.