OpenAI has unveiled GPT-5.2, its most advanced AI model to date, promising meaningful improvements for the hundreds of millions of people who rely on ChatGPT each week. The new model is rolling out in ChatGPT now and is also available to developers through the OpenAI API, marking a significant step forward for both everyday users and businesses that depend on AI for real work.
With more than 800 million weekly ChatGPT users and over 4 million developers on the OpenAI platform, GPT-5.2 is designed to make the AI feel noticeably more capable, more reliable and far more helpful in handling complex professional tasks. OpenAI says the flagship version, GPT-5.2 Thinking, has reached “expert-level” performance on GDPval, the company’s benchmark that evaluates how well AI handles 44 real-world professional tasks such as creating spreadsheets, drafting documents and building presentations. In company testing, the new model not only outperforms its predecessors but in many cases matches or surpasses the results produced by human professionals.
GPT-5.2 arrives in three variants aimed at different types of tasks. The standard GPT-5.2 model, often referred to as “Instant,” focuses on speed and clarity, making it better at everyday information-seeking, troubleshooting, writing assistance, translation and study support. The new “Thinking” version is engineered for deeper reasoning and more advanced work, with stronger problem-solving, better accuracy in formatting and structuring spreadsheets, and early improvements in building slide decks from scratch. At the top end is GPT-5.2 Pro, a slower but more meticulous model designed for complex technical questions, intensive research and professional-grade programming support. OpenAI says Pro offers improved accuracy, fewer major errors and consistently stronger performance in complex domains where high-quality answers matter most.
PRACTICAL IMPROVEMENTS
For developers, GPT-5.2 represents OpenAI’s strongest platform yet for building AI agents. The company reports significant gains in general intelligence, long-context understanding, vision capabilities and tool-calling — the feature that allows the model to control external systems or take multi-step actions. These upgrades are aimed at enabling developers to build agents that can execute complicated tasks end-to-end with far greater reliability, from processing long documents to navigating multi-part workflows that previously required human oversight.
Across the board, GPT-5.2 also brings practical improvements that users will notice in day-to-day work. It handles long documents more gracefully, maintains context over extended conversations, produces more accurate code, interprets charts and screenshots with higher precision and generally offers a more polished, less error-prone experience. For businesses and professionals who have already been using AI to cut down on administrative burdens, OpenAI says GPT-5.2 should deliver another meaningful boost in productivity.
The launch arrives at a time when competition in generative AI is more intense than ever, with rivals like Google and Anthropic rapidly advancing their own models. OpenAI’s focus with GPT-5.2 is squarely on deepening ChatGPT’s usefulness for real work and positioning the service as a dependable everyday assistant rather than just a clever conversational tool.
ROLLING OUT NOW
GPT-5.2 is now rolling out to paid ChatGPT users, including Plus, Pro, Business and Enterprise customers, while developers can access the new capabilities through the API immediately. For the millions already using ChatGPT as part of their working day, the upgrade aims to make the AI feel less like a novelty and more like a genuine partner in getting things done.
