Microsoft continues to strengthen Copilot with the integration of GPT-5.2, a free update now rolling out across Copilot Web, Windows, and mobile devices. It will coexist with the GPT-5.1 model to provide greater flexibility depending on the task at hand.
Considered by OpenAI to be its most powerful model family to date, GPT-5.2 boosts Copilot’s performance in real-world business scenarios, enabling the execution of complex tasks in less time.
From creating spreadsheets and presentations, to writing and reviewing code, analyzing lengthy documents, using tools intelligently, and working with images—this new version reinforces Microsoft Copilot as a strategic ally for enhancing productivity and operational efficiency.
GPT 5.2 Plus in Copilot
Last month, Microsoft integrated GPT-5.1 into Copilot through Smart Mode, improving automatic model selection based on the task. Now, the company is taking it a step further with GPT-5.2, introduced as the new “Smart Plus” mode, designed for scenarios that demand greater analytical capability and precision.
In Copilot, GPT-5.2 is positioned as a “Thinking” variant, optimized to handle complex, high-value tasks—especially in professional settings. Microsoft highlights that this model delivers superior performance when it comes to deep analysis, planning, and the generation of advanced deliverables.
In knowledge work evaluations (GDPval) across 44 occupations, GPT-5.2 Thinking matches or outperforms industry professionals in 70.9% of cases, compared to 38.8% for GPT-5. This gap marks a significant leap in AI’s ability to support critical business functions.
This performance jump explains why OpenAI defines GPT-5.2 as an “expert-level” model for many clearly structured office tasks, such as creating presentations, schedules, work models, and other key deliverables.
In the realm of software development, GPT-5.2 Thinking also shows a clear improvement: scoring 55.6% on SWE-Bench Pro and 80% on SWE-Bench Verified—outperforming GPT-5.1 Thinking and solidifying itself as a more reliable tool for programming and code review tasks.
The arrival of GPT-5.2 marks a meaningful evolution from GPT-5.1, particularly in professional and development environments. This new series introduces deeper chains of reasoning, significantly richer context handling, and agentic execution capabilities that enable the generation of ready-to-use deliverables with fewer iterations.
In practice, GPT-5.2 streamlines the creation of design documents, executable code, unit tests, and deployment scripts—faster and more consistently—reducing friction in technical workflows. This is made possible by a new architecture designed for higher performance, efficiency, and reasoning depth compared to previous generations.
Additionally, GPT-5.2 builds on the validated dataset of GPT-5.1, incorporating improvements in safety, stability, and integrations—resulting in more reliable outputs and stronger performance across key metrics.
With the launch of GPT-5.2 and GPT-5.2-Chat, Microsoft and OpenAI are doubling down on AI built for daily operational excellence. Both models offer clear upgrades over their predecessors and address different needs within the same ecosystem.
GPT-5.2 stands as the most advanced reasoning model, capable of solving complex problems more efficiently and with greater refinement. For informational and analytical tasks, it combines stronger reasoning with enhanced communication skills, delivering clearer, more structured results in spreadsheets, reports, and presentations.
Meanwhile, GPT-5.2-Chat functions as a powerful, versatile model optimized for daily work and learning. It offers notable improvements in information retrieval, tutorials and guides, technical writing, and translation, while also being more effective as a support tool for training, skill development, and career guidance.
Together, these advancements position GPT-5.2 as a more mature, secure, and enterprise-ready AI platform—aligned with today’s demands for efficiency, precision, and scalability.
Read more: What is Microsoft Copilot in Azure?
For businesses to scale the use of artificial intelligence in complex professional tasks, fast responses alone aren’t enough. What’s required are structured results, reliable tool usage, and enterprise-grade security controls.
In this context, GPT-5.2 emerges as a new benchmark—optimized for agentic scenarios within the Microsoft Foundry enterprise platform, delivering a consistent experience for developers across reasoning, chat, and programming.
Multi-step Reasoning Chains: GPT-5.2 breaks down complex problems into logical steps, justifies decisions, and generates explainable plans—critical in environments where traceability and validation are essential.
Context-Aware Planning: The model can process large volumes of information (such as project summaries, codebases, or meeting notes) to produce holistic, coherent, and actionable outcomes aligned with business objectives.
End-to-End Agentic Execution: GPT-5.2 coordinates complete workflows—including design, implementation, testing, and deployment—significantly reducing iteration cycles and the need for manual oversight.
Enterprise Security and Governance: It features advanced controls, managed identities, and policy enforcement—enabling secure and compliant AI adoption, a must-have in regulated corporate environments.
Thanks to its advanced reasoning, greater context capacity, and mature agentic patterns, GPT-5.2 is a strategic choice for developing AI agents capable of handling complex, long-duration tasks across various industries, such as:
Financial Services
Healthcare
Manufacturing
Customer Support
Decision Support and Analysis: Ideal for simulations, evaluating alternatives, and creating defensible plans for stakeholders, with clear explanations of trade-offs and risks.
Application Modernization: Speeds up refactoring processes, test generation, and migration planning—incorporating risk criteria and rollback strategies.
Data Management and Pipelines: Audits ETL processes, recommends monitors and SLAs, and generates validation SQL to ensure data integrity and quality.
Customer Experiences: Enables the creation of contextual assistants and agentic workflows that integrate natively into existing applications, enhancing the end-user experience.
The result is reliable AI agents capable of maintaining consistency across complex workflows and interconnected services, while producing structured, auditable, and scalable outcomes. All of this positions GPT-5.2 as a solid foundation for driving secure, sustainable enterprise AI adoption within Microsoft Foundry.