What are Foundry Tools?
Foundry Tools is Microsoft’s current collective name for prebuilt, cloud-based AI services that let developers integrate cognitive capabilities into applications without needing to build their own AI or data science expertise. The term replaces the earlier designation “Azure AI Services” or “Cognitive Services”; the underlying APIs and SKUs generally remain unchanged.
Active Foundry Tools include Speech (speech-to-text, text-to-speech, speaker recognition), Translator (translation), Language (text understanding), Vision (image and video analysis), Document Intelligence (document extraction), Content Understanding (analysis of various media types), Content Safety (detection of harmful content), Custom Vision (custom image classification training), and Immersive Reader. Most services are accessible via REST APIs and client SDKs in common programming languages and can be tried out in the Foundry portal. Azure AI Search is also available as a tightly integrated search service for AI applications.
Some earlier Cognitive Services offerings have been retired: Anomaly Detector, Content Moderator, Language Understanding (LUIS), Metrics Advisor, Personalizer, and QnA Maker are retired and remain available only for existing applications. For new projects, Microsoft recommends the respective successor services.
Core Features
- Broad portfolio of prebuilt AI APIs for language, vision, documents, translation, and content safety
- Free F0 tiers for testing and production-ready S0 tiers with higher rate limits
- Access via REST APIs, client SDKs, and interactive web interfaces (e.g., Content Safety Studio, Custom Vision portal)
- Container support for many services for on-premises or edge deployment
- Security model with Microsoft Entra authentication, resource keys, and virtual networks
- Combinable with Foundry Agent Service and Foundry Observability to build complete AI applications
Typical Use Cases
Organizations integrate Speech and Translator APIs to build multilingual customer service applications with transcription and live translation, without training their own speech recognition models.
Document-intensive industries such as financial services, insurance, and legal use Document Intelligence and Content Understanding to automatically turn forms, contracts, and other documents into structured data.
Teams building their own LLM applications or agents use Content Safety as a guardrail layer and combine Vision or Language APIs with Azure AI Search for RAG scenarios.
Retail and manufacturing use Vision and Custom Vision capabilities for image analysis, such as product recognition or quality control.
Benefits
- Fast entry into AI functionality without training your own models
- Consistent security, compliance, and billing mechanisms across all services
- Seamless integration with Foundry Agent Service and Foundry Observability for more complex AI applications
- Flexible deployment options from cloud to container/on-premises
- Broad language and regional coverage for global applications
Integration with innFactory
As a Microsoft Solutions Partner, innFactory supports you in selecting, integrating, and operating the right Foundry Tools for your use case: from architecture through proof of concepts to production integration into your existing system landscape.
Contact us for a non-binding consultation on Foundry Tools and Microsoft Azure.
Typical Use Cases
Frequently Asked Questions
What are Foundry Tools?
Foundry Tools is Microsoft's current brand name for the collection of prebuilt, cloud-based AI APIs and models (formerly Azure AI Services or Cognitive Services). It includes services such as Speech, Language, Vision, Document Intelligence, Content Safety, Translator, Content Understanding, Custom Vision, and Immersive Reader.
How do Foundry Tools differ from Foundry Agent Service?
Foundry Tools are individually usable AI APIs for specific tasks such as speech recognition, image analysis, or document extraction. Foundry Agent Service is the platform for building autonomous AI agents that orchestrate and combine these tools along with models from the Foundry model catalog.
How are Foundry Tools billed?
Billing is generally usage-based per transaction, tiered by pricing plan (e.g., F0 for a free entry point, S0 for production use with higher rate limits). Details vary by service and are listed on the respective pricing pages.
Are any Foundry Tools being retired?
Yes, several earlier Cognitive Services offerings such as Anomaly Detector, Content Moderator, Language Understanding (LUIS), Metrics Advisor, Personalizer, and QnA Maker have been retired. They remain available for existing applications but should not be used for new projects.
Can I run Foundry Tools on-premises or in my own data center?
Many Foundry Tools can be deployed via containers for on-premises operation, for example for compliance or data protection reasons. The feature set of the containers can differ from the full cloud version.
Note: All product information on this page has been compiled with care, but is provided without guarantee and may be outdated or incomplete. Cloud services evolve rapidly — features, pricing, SLAs, and availability change frequently. Authoritative and up-to-date information can only be found on the official product page of Azure (official documentation). This page does not represent an offer by Azure.
