Business Software for Microsoft Foundry

Confidently advance the future of safe and ethical AI applications within your organization. Utilize cutting-edge technologies and established best practices to effectively scale AI while managing risks, enhancing accuracy, safeguarding privacy, ensuring transparency, and streamlining compliance efforts. Equip cross-functional teams with the necessary resources to create the next wave of AI applications in a secure manner, leveraging integrated tools and templates designed to incorporate responsible AI into open source, machine learning operations, and generative AI processes. Proactively identify and address potential misuse through robust responsible AI measures, top-tier Azure security, and specialized AI tools. Monitor both text and images to swiftly recognize and filter out offensive or inappropriate content. Accelerate the deployment of machine learning models and foster collaboration through prompt flow, ultimately achieving a faster return on investment. Build innovative generative AI applications and tailor-made copilots all within a single, cohesive platform, ensuring efficiency and effectiveness in your AI initiatives. Through these strategies, you can create a safer AI landscape that not only meets regulatory requirements but also builds trust with users and stakeholders alike.

Azure AI Content Understanding empowers organizations to convert unstructured multimodal data into actionable insights. By extracting valuable information from various input formats including text, audio, images, and video, businesses can unlock essential insights. Employing advanced AI techniques like schema extraction and grounding, it ensures the generation of accurate, high-quality data suitable for further applications. This technology simplifies the integration of diverse data types into a cohesive workflow, resulting in reduced costs and an expedited path to value realization. For instance, businesses and call center operators can leverage insights from call recordings to monitor crucial KPIs, improve product experiences, and respond to customer inquiries more efficiently and accurately. Furthermore, by ingesting a wide array of data types such as documents, images, audio, or video, organizations can utilize various AI models offered in Azure AI to convert raw input into structured outputs that facilitate easier processing and analysis in subsequent applications. Such capabilities ultimately enhance decision-making processes across various sectors.

Phi-4 is an advanced small language model (SLM) comprising 14 billion parameters, showcasing exceptional capabilities in intricate reasoning tasks, particularly in mathematics, alongside typical language processing functions. As the newest addition to the Phi family of small language models, Phi-4 illustrates the potential advancements we can achieve while exploring the limits of SLM technology. It is currently accessible on Azure AI Foundry under a Microsoft Research License Agreement (MSRLA) and is set to be released on Hugging Face in the near future. Due to significant improvements in processes such as the employment of high-quality synthetic datasets and the careful curation of organic data, Phi-4 surpasses both comparable and larger models in mathematical reasoning tasks. This model not only emphasizes the ongoing evolution of language models but also highlights the delicate balance between model size and output quality. As we continue to innovate, Phi-4 stands as a testament to our commitment to pushing the boundaries of what's achievable within the realm of small language models.

Microsoft has introduced Muse, an innovative generative AI model poised to transform the way gameplay concepts are developed. In partnership with Ninja Theory, this World and Human Action Model (WHAM) draws training data from the game Bleeding Edge, granting it a profound grasp of 3D game landscapes, including the intricacies of physics and player interactions. This capability allows Muse to generate varied and coherent gameplay sequences, which can enhance the creative process for developers. Additionally, the AI is capable of creating game visuals and anticipating controller actions, streamlining prototyping and artistic exploration in game design. By leveraging an analysis of over 1 billion images and actions, Muse showcases its potential not only for game creation but also for game preservation, as it can recreate classic titles for contemporary gaming platforms. Despite being in its initial phases, with output currently limited to a resolution of 300×180 pixels, Muse signifies a pivotal step forward in harnessing AI to support game development, with the goal of amplifying human creativity rather than supplanting it. As Muse evolves, it may open up new avenues for both game innovation and the revival of beloved gaming classics.

Magma is an advanced AI model designed to seamlessly integrate digital and physical environments, offering both vision-language understanding and the ability to perform actions in both realms. By pretraining on large, diverse datasets, Magma enhances its capacity to handle a wide variety of tasks that require spatial intelligence and verbal understanding. Unlike previous Vision-Language-Action (VLA) models that are limited to specific tasks, Magma is capable of generalizing across new environments, making it an ideal solution for creating AI assistants that can interact with both software interfaces and physical objects. It outperforms specialized models in UI navigation and robotic manipulation tasks, providing a more adaptable and capable AI agent.

Phi-4-reasoning is an advanced transformer model featuring 14 billion parameters, specifically tailored for tackling intricate reasoning challenges, including mathematics, programming, algorithm development, and strategic planning. Through a meticulous process of supervised fine-tuning on select "teachable" prompts and reasoning examples created using o3-mini, it excels at generating thorough reasoning sequences that optimize computational resources during inference. By integrating outcome-driven reinforcement learning, Phi-4-reasoning is capable of producing extended reasoning paths. Its performance notably surpasses that of significantly larger open-weight models like DeepSeek-R1-Distill-Llama-70B and nears the capabilities of the comprehensive DeepSeek-R1 model across various reasoning applications. Designed for use in settings with limited computing power or high latency, Phi-4-reasoning is fine-tuned with synthetic data provided by DeepSeek-R1, ensuring it delivers precise and methodical problem-solving. This model's ability to handle complex tasks with efficiency makes it a valuable tool in numerous computational contexts.

Phi-4-reasoning-plus is an advanced reasoning model with 14 billion parameters, enhancing the capabilities of the original Phi-4-reasoning. It employs reinforcement learning for better inference efficiency, processing 1.5 times the number of tokens compared to its predecessor, which results in improved accuracy. Remarkably, this model performs better than both OpenAI's o1-mini and DeepSeek-R1 across various benchmarks, including challenging tasks in mathematical reasoning and advanced scientific inquiries. Notably, it even outperforms the larger DeepSeek-R1, which boasts 671 billion parameters, on the prestigious AIME 2025 assessment, a qualifier for the USA Math Olympiad. Furthermore, Phi-4-reasoning-plus is accessible on platforms like Azure AI Foundry and HuggingFace, making it easier for developers and researchers to leverage its capabilities. Its innovative design positions it as a top contender in the realm of reasoning models.

Phi-4-mini-reasoning is a transformer-based language model with 3.8 billion parameters, specifically designed to excel in mathematical reasoning and methodical problem-solving within environments that have limited computational capacity or latency constraints. Its optimization stems from fine-tuning with synthetic data produced by the DeepSeek-R1 model, striking a balance between efficiency and sophisticated reasoning capabilities. With training that encompasses over one million varied math problems, ranging in complexity from middle school to Ph.D. level, Phi-4-mini-reasoning demonstrates superior performance to its base model in generating lengthy sentences across multiple assessments and outshines larger counterparts such as OpenThinker-7B, Llama-3.2-3B-instruct, and DeepSeek-R1. Equipped with a 128K-token context window, it also facilitates function calling, which allows for seamless integration with various external tools and APIs. Moreover, Phi-4-mini-reasoning can be quantized through the Microsoft Olive or Apple MLX Framework, enabling its deployment on a variety of edge devices, including IoT gadgets, laptops, and smartphones. Its design not only enhances user accessibility but also expands the potential for innovative applications in mathematical fields.

Foundry Local serves as a localized iteration of Azure AI Foundry, allowing users to run large language models (LLMs) directly on their Windows machines. This AI inference solution, executed on-device, ensures enhanced privacy, tailored customization, and financial advantages over cloud-based services. Furthermore, it seamlessly integrates into your current workflows and applications, offering a straightforward command-line interface (CLI) and REST API for user convenience. This makes it an ideal choice for those seeking to leverage AI capabilities while maintaining control over their data.

Magistral is the inaugural language model family from Mistral AI that emphasizes reasoning, offered in two variants: Magistral Small, a 24 billion parameter open-weight model accessible under Apache 2.0 via Hugging Face, and Magistral Medium, a more robust enterprise-grade version that can be accessed through Mistral's API, the Le Chat platform, and various major cloud marketplaces. Designed for specific domains, it excels in transparent, multilingual reasoning across diverse tasks such as mathematics, physics, structured calculations, programmatic logic, decision trees, and rule-based systems, generating outputs that follow a chain of thought in the user's preferred language, which can be easily tracked and validated. This release signifies a transition towards more compact yet highly effective transparent AI reasoning capabilities. Currently, Magistral Medium is in preview on platforms including Le Chat, the API, SageMaker, WatsonX, Azure AI, and Google Cloud Marketplace. Its design is particularly suited for general-purpose applications that necessitate extended thought processes and improved accuracy compared to traditional non-reasoning language models. The introduction of Magistral represents a significant advancement in the pursuit of sophisticated reasoning in AI applications.

Phi-4-mini-flash-reasoning is a 3.8 billion-parameter model that is part of Microsoft's Phi series, specifically designed for edge, mobile, and other environments with constrained resources where processing power, memory, and speed are limited. This innovative model features the SambaY hybrid decoder architecture, integrating Gated Memory Units (GMUs) with Mamba state-space and sliding-window attention layers, achieving up to ten times the throughput and a latency reduction of 2 to 3 times compared to its earlier versions without compromising on its ability to perform complex mathematical and logical reasoning. With a support for a context length of 64K tokens and being fine-tuned on high-quality synthetic datasets, it is particularly adept at handling long-context retrieval, reasoning tasks, and real-time inference, all manageable on a single GPU. Available through platforms such as Azure AI Foundry, NVIDIA API Catalog, and Hugging Face, Phi-4-mini-flash-reasoning empowers developers to create applications that are not only fast but also scalable and capable of intensive logical processing. This accessibility allows a broader range of developers to leverage its capabilities for innovative solutions.

gpt-oss-20b is a powerful text-only reasoning model consisting of 20 billion parameters, made available under the Apache 2.0 license and influenced by OpenAI’s gpt-oss usage guidelines, designed to facilitate effortless integration into personalized AI workflows through the Responses API without depending on proprietary systems. It has been specifically trained to excel in instruction following and offers features like adjustable reasoning effort, comprehensive chain-of-thought outputs, and the ability to utilize native tools such as web search and Python execution, resulting in structured and clear responses. Developers are responsible for establishing their own deployment precautions, including input filtering, output monitoring, and adherence to usage policies, to ensure that they align with the protective measures typically found in hosted solutions and to reduce the chance of malicious or unintended actions. Additionally, its open-weight architecture makes it particularly suitable for on-premises or edge deployments, emphasizing the importance of control, customization, and transparency to meet specific user needs. This flexibility allows organizations to tailor the model according to their unique requirements while maintaining a high level of operational integrity.

gpt-oss-120b is a text-only reasoning model with 120 billion parameters, released under the Apache 2.0 license and managed by OpenAI’s usage policy, developed with insights from the open-source community and compatible with the Responses API. It is particularly proficient in following instructions, utilizing tools like web search and Python code execution, and allowing for adjustable reasoning effort, thereby producing comprehensive chain-of-thought and structured outputs that can be integrated into various workflows. While it has been designed to adhere to OpenAI's safety policies, its open-weight characteristics present a risk that skilled individuals might fine-tune it to circumvent these safeguards, necessitating that developers and enterprises apply additional measures to ensure safety comparable to that of hosted models. Evaluations indicate that gpt-oss-120b does not achieve high capability thresholds in areas such as biological, chemical, or cyber domains, even following adversarial fine-tuning. Furthermore, its release is not seen as a significant leap forward in biological capabilities, marking a cautious approach to its deployment. As such, users are encouraged to remain vigilant about the potential implications of its open-weight nature.

Claude Opus 4.1 represents a notable incremental enhancement over its predecessor, Claude Opus 4, designed to elevate coding, agentic reasoning, and data-analysis capabilities while maintaining the same level of deployment complexity. This version boosts coding accuracy to an impressive 74.5 percent on SWE-bench Verified and enhances the depth of research and detailed tracking for agentic search tasks. Furthermore, GitHub has reported significant advancements in multi-file code refactoring, and Rakuten Group emphasizes its ability to accurately identify precise corrections within extensive codebases without introducing any bugs. Independent benchmarks indicate that junior developer test performance has improved by approximately one standard deviation compared to Opus 4, reflecting substantial progress consistent with previous Claude releases.

Claude Sonnet 4.5 represents Anthropic's latest advancement in AI, crafted to thrive in extended coding environments, complex workflows, and heavy computational tasks while prioritizing safety and alignment. It sets new benchmarks with its top-tier performance on the SWE-bench Verified benchmark for software engineering and excels in the OSWorld benchmark for computer usage, demonstrating an impressive capacity to maintain concentration for over 30 hours on intricate, multi-step assignments. Enhancements in tool management, memory capabilities, and context interpretation empower the model to engage in more advanced reasoning, leading to a better grasp of various fields, including finance, law, and STEM, as well as a deeper understanding of coding intricacies. The system incorporates features for context editing and memory management, facilitating prolonged dialogues or multi-agent collaborations, while it also permits code execution and the generation of files within Claude applications. Deployed at AI Safety Level 3 (ASL-3), Sonnet 4.5 is equipped with classifiers that guard against inputs or outputs related to hazardous domains and includes defenses against prompt injection, ensuring a more secure interaction. This model signifies a significant leap forward in the intelligent automation of complex tasks, aiming to reshape how users engage with AI technologies.

Microsoft Copilot in Azure redefines cloud operations with a generative AI companion built directly into the Azure platform. Designed for developers, IT admins, and cloud architects, it leverages natural language to help users configure, deploy, and manage resources with precision. By analyzing your environment, Copilot provides tailored recommendations for scalability, reliability, cost optimization, and security compliance. Its intelligent automation assists with complex commands, real-time troubleshooting, and data orchestration across multiple Azure services. With deep integration into tools like Azure Monitor, Kubernetes Service, and Backup, Copilot ensures a seamless workflow from application development to infrastructure management. Microsoft’s robust security framework—with 34,000 dedicated engineers and 100+ certifications—ensures every AI-assisted action meets enterprise standards. The result is faster problem-solving, improved efficiency, and reduced operational overhead. Built on responsible AI principles, Copilot in Azure combines innovation and trust to help organizations maximize the value of their cloud investments.

GPT-5-Codex-Mini provides a more resource-efficient way to code, allowing approximately four times the usage compared to GPT-5-Codex while maintaining dependable functionality for most development needs. It performs exceptionally well for straightforward coding, automation, and maintenance tasks where full-scale model power isn’t required. Integrated into the CLI and IDE extension via ChatGPT sign-in, it’s designed for accessibility and convenience across environments. When users approach 90% of their rate limits, the system proactively recommends switching to the Mini model to ensure continuous workflow. ChatGPT Plus, Business, and Edu accounts enjoy 50% higher rate limits, giving developers more capacity for sustained sessions. Pro and Enterprise plans gain priority processing, making response times noticeably faster during peak usage. The overall system architecture has been optimized for GPU efficiency, contributing to higher throughput and reduced latency. Together, these refinements make Codex more versatile and reliable for both individual and professional programming work.

GPT-5.1 Instant is an advanced AI model tailored for everyday users, merging rapid response times with enhanced conversational warmth. Its adaptive reasoning capability allows it to determine the necessary computational effort for tasks, ensuring swift responses while maintaining a deep level of understanding. By focusing on improved instruction adherence, users can provide detailed guidance and anticipate reliable execution. Additionally, the model features expanded personality controls, allowing the chat tone to be adjusted to Default, Friendly, Professional, Candid, Quirky, or Efficient, alongside ongoing trials of more nuanced voice modulation. The primary aim is to create interactions that feel more organic and less mechanical, all while ensuring robust intelligence in writing, coding, analysis, and reasoning tasks. Furthermore, GPT-5.1 Instant intelligently manages user requests through the main interface, deciding whether to employ this version or the more complex “Thinking” model based on the context of the query. Ultimately, this innovative approach enhances user experience by making interactions more engaging and tailored to individual preferences.

GPT-5.1 Thinking represents an evolved reasoning model within the GPT-5.1 lineup, engineered to optimize "thinking time" allocation according to the complexity of prompts, allowing for quicker responses to straightforward inquiries while dedicating more resources to tackle challenging issues. In comparison to its earlier version, it demonstrates approximately double the speed on simpler tasks and takes twice as long for more complex ones. The model emphasizes clarity in its responses, minimizing the use of jargon and undefined terminology, which enhances the accessibility and comprehensibility of intricate analytical tasks. It adeptly modifies its reasoning depth, ensuring a more effective equilibrium between rapidity and thoroughness, especially when addressing technical subjects or multi-step inquiries. By fusing substantial reasoning power with enhanced clarity, GPT-5.1 Thinking emerges as an invaluable asset for handling complicated assignments, including in-depth analysis, programming, research, or technical discussions, while simultaneously decreasing unnecessary delays for routine requests. This improved efficiency not only benefits users seeking quick answers but also supports those engaged in more demanding cognitive tasks.

Agent 365 is Microsoft’s new enterprise framework for managing AI agents with the same rigor and structure used for human users. It centralizes oversight by providing a registry that surfaces every agent operating within your organization, including identity-secured agents, internally registered agents, and automatically detected shadow agents. The platform enhances security by extending Microsoft Defender protections, Entra identity access controls, and Purview governance policies to all agents. Agent 365 integrates with Microsoft 365, Power Apps, Power Automate, and Power BI, enabling agents to participate in workflows, analytics, and productivity tasks just like any other digital worker. Using Work IQ, organizations can equip agents with deep contextual understanding sourced from company data, relationships, and internal systems. This unified approach simplifies deployment, strengthens compliance, and improves operational insight for IT teams. Through Microsoft’s Frontier early access program, IT admins can explore and activate Agent 365 directly in the Admin Center. Microsoft built Agent 365 to support the rapidly growing role of AI agents across enterprise environments, ensuring they remain secure, governed, and aligned with organizational standards.

Anthropic’s release of Claude Opus 4.5 introduces a frontier AI model that excels at coding, complex reasoning, deep research, and long-context tasks. It sets new performance records on real-world engineering benchmarks, handling multi-system debugging, ambiguous instructions, and cross-domain problem solving with greater precision than earlier versions. Testers and early customers reported that Opus 4.5 “just gets it,” offering creative reasoning strategies that even benchmarks fail to anticipate. Beyond raw capability, the model brings stronger alignment and safety, with notable advances in prompt-injection resistance and behavior consistency in high-stakes scenarios. The Claude Developer Platform also gains richer controls including effort tuning, multi-agent orchestration, and context management improvements that significantly boost efficiency. Claude Code becomes more powerful with enhanced planning abilities, multi-session desktop support, and better execution of complex development workflows. In the Claude apps, extended memory and automatic context summarization enable longer, uninterrupted conversations. Together, these upgrades showcase Opus 4.5 as a highly capable, secure, and versatile model designed for both professional workloads and everyday use.

GPT-5.2 marks a new milestone in the evolution of the GPT-5 series, bringing heightened intelligence, richer context understanding, and smoother conversational behavior. The updated architecture introduces multiple enhanced variants that work together to produce clearer reasoning and more accurate interpretations of user needs. GPT-5.2 Instant remains the main model for everyday interactions, now upgraded with faster response times, stronger instruction adherence, and more reliable contextual continuity. For users tackling complex or layered tasks, GPT-5.2 Thinking provides deeper cognitive structure, offering step-by-step explanations, stronger logical flow, and improved endurance across long-form reasoning challenges. The platform automatically determines which model variant is optimal for any query, ensuring users always benefit from the most appropriate capabilities. These advancements reduce friction, simplify workflows, and produce answers that feel more grounded and intention-aware. In addition to intelligence upgrades, GPT-5.2 emphasizes conversational naturalness, making exchanges feel more intuitive and humanlike. Overall, this release delivers a more capable, responsive, and adaptive AI experience across all forms of interaction.

Grok 4.1 Thinking is the reasoning-enabled version of Grok designed to handle complex, high-stakes prompts with deliberate analysis. Unlike fast-response models, it visibly works through problems using structured reasoning before producing an answer. This approach improves accuracy, reduces misinterpretation, and strengthens logical consistency across longer conversations. Grok 4.1 Thinking leads public benchmarks in general capability and human preference testing. It delivers advanced performance in emotional intelligence by understanding context, tone, and interpersonal nuance. The model is especially effective for tasks that require judgment, explanation, or synthesis of multiple ideas. Its reasoning depth makes it well-suited for analytical writing, strategy discussions, and technical problem-solving. Grok 4.1 Thinking also demonstrates strong creative reasoning without sacrificing coherence. The model maintains alignment and reliability even in ambiguous scenarios. Overall, it sets a new standard for transparent and thoughtful AI reasoning.

GPT-5.2-Codex is a next-generation coding model created to support advanced, agent-driven software development. Built on the GPT-5.2 architecture, it is fine-tuned specifically for real-world engineering tasks. The model excels at working across large codebases while preserving context over long sessions. It handles complex refactors, migrations, and multi-step implementations more reliably than previous Codex models. GPT-5.2-Codex demonstrates top-tier performance in realistic terminal environments. Enhanced tool-calling and improved factual accuracy make it suitable for production workflows. The model is also significantly stronger in cybersecurity-related tasks. It can assist with vulnerability research and defensive security analysis. GPT-5.2-Codex includes safeguards designed to support responsible deployment. It represents a major advancement in professional-grade coding AI.

GPT-5.3-Codex is a next-generation AI agent built to expand Codex beyond code writing into full-spectrum professional execution. It unifies advanced coding intelligence with reasoning, planning, and computer-use capabilities. The model delivers faster performance while handling more complex workflows across development environments. GPT-5.3-Codex can autonomously iterate on large projects while remaining interactive and steerable. It supports tasks such as debugging, deployment, performance optimization, and system monitoring. The model demonstrates state-of-the-art results across real-world coding benchmarks. It also excels at web development, generating production-ready applications from minimal prompts. GPT-5.3-Codex understands intent more effectively, producing stronger default designs and functionality. Its agentic nature allows it to operate like a collaborative teammate. This makes it suitable for both individual developers and large teams.