OpenAI Releases GPT-5.2 Thinking with Large Benchmark Gains Over Prior Models

Tweet lists GPT-5.2 Thinking benchmark improvements: SWE-Bench 55.6% (from 50.8), GPQA 92.4% (from 88.1), CharXiv 82.1% (from 67.0), FrontierMath 40.3% (from 31.0), AIME 100% (from 94%), ARC-AGI1 86.2% (from 72.8), ARC-AGI2 52.9% (from 17.6), GDPval 70.9% (from 38.8), with claim of large overall gains.