🤯GPT-5.2: The Future of Work Arrives! 🚀

Tech

🎧

Microsoft Unveils GPT-5.2 “Smart Plus” – A Leap in AI Capability
Microsoft Copilot is introducing GPT 5.2 as “Smart Plus” mode, representing OpenAI’s most advanced model series to date. This new mode enables users to complete real tasks more quickly, including building spreadsheets and presentations, writing and reviewing code, understanding lengthy documents, utilizing various tools, and working with images.

GPT-5.2 Outperforms GPT-5 by a Wide Margin
Benchmarking across “knowledge work” tasks (GDPval, encompassing 44 occupations) revealed that GPT-5.2 Thinking outperformed GPT-5 in 70.9% of instances, compared to 38.8% for GPT-5. This significant improvement has led OpenAI to characterize GPT-5.2 as “expert-level” for many specific office deliverables, such as decks, schedules, and models.

Coding Prowess: GPT-5.2 Achieves New Heights
Furthermore, GPT-5.2 Thinking achieved notable results on coding benchmarks, posting 55.6% on SWE-Bench Pro and 80% on SWE-bench Verified—substantially higher than GPT-5.1 Thinking.

Across Diverse Assessments, GPT-5.2 Continues to Lead
Across a range of evaluations, including GPQA Diamond (92.4%), AIME 2025 (100%), CharXiv Reasoning (with Python) (88.7%), and ARC-AGI (including a notable increase on ARC-AGI-2), GPT 5.2 consistently demonstrated superior performance.

This article is AI-synthesized from public sources and may not reflect original reporting.