OpenAI has launched GPT-5.2, its most advanced model for professional tasks. Users already report significant time savings, averaging 40 to 60 minutes daily, with heavy users gaining over 10 hours weekly. GPT-5.2 enhances productivity by improving capabilities in areas like spreadsheet creation, presentation building, code writing, and managing complex projects. In benchmarking tests, it outperformed professionals in 44 job roles, achieving a striking 70.9% success rate in knowledge work tasks compared to the 38.8% achieved by GPT-5.1. The model excels in long-context reasoning, achieving top scores in the OpenAI MRCRv2 evaluation. It can accurately process extensive documents, making it useful for professionals handling reports, contracts, and research papers. Notably, it reached near 100% accuracy on a specific evaluation variant, demonstrating its strength in maintaining coherence across lengthy texts. GPT-5.2 also integrates a new Responses `/compact` endpoint, which extends context limits, allowing for more complex, multi-step workflows. In terms of visual tasks, GPT-5.2 cuts error rates significantly for chart reasoning and interpreting software interfaces. It can better analyze visual data, aiding workflows in finance, engineering, and design. For instance, it shows improved spatial awareness in identifying components within images, outperforming its predecessor by a notable margin. Overall, GPT-5.2 brings substantial advancements in general intelligence and practical applications, making it a valuable tool for various professional settings.