OpenAI says its new GPT 5.2 set a 'new state-of-the-art score' for professional knowledge work

"the company said in a statement. In a benchmark test called GDPval, OpenAI said its new AI model can outperform "industry professionals at well-specified knowledge work tasks spanning 44 occupations." "GPT-5.2 Thinking produced outputs for GDPval tasks at >11x the speed and <1% the cost of expert professionals, suggesting that when paired with human oversight, GPT-5.2 can help with professional work,""

"And in a note that is sure to catch the attention of bankers, OpenAI wrote that in an internal benchmark of junior investment banking analyst spreadsheet modeling tasks - "such as putting together a three-statement model for a Fortune 500 company with proper formatting and citations, or building a leveraged buyout model for a take-private" - the new model's score per task was "9.3% higher than GPT-5.1's, rising from 59.1% to 68.4%" on average."

GPT-5.2 enhances capabilities for professional knowledge work by improving spreadsheet creation, presentation building, code writing, image perception, long-context understanding, tool use, and complex multi-step project handling. GDPval benchmarking indicates outperformance of industry professionals across 44 occupations, with GPT-5.2 Thinking producing outputs at over 11x the speed and under 1% of expert cost when paired with human oversight. Internal investment-banking modeling benchmarks show average task scores rose from 59.1% to 68.4%, a 9.3% improvement over GPT-5.1. The model also shows gains in agentic coding ability amid increased industry competition.

Read at Business Insider