Hosted on MSN

OpenAI’s GPT-5.5 just posted a massive jump in math and multimodal reasoning — scoring 81 on a test the old model routinely failed

When researchers at Tsinghua University and other institutions built MMMU-Pro, they designed it to be nearly impossible to game. Every question pairs an image with text, and any item a model can ...
ZDNet

I put GPT-5.5 through a 10-round test: It scored 93/100, losing points only for exuberance

GPT-5.5 delivers polished, useful answers across tasks. Strong performance across writing, coding, and reasoning tasks. Overeagerness hurts accuracy and instruction following. OpenAI has released ...
Rolling Out

OpenAI’s GPT-5.5 Instant reduces hallucinations by 52.5%

ChatGPT just got a quieter but significant upgrade, and most users will feel it before they even notice it. OpenAI rolled out GPT-5.5 Instant on Tuesday as the new default model for all ChatGPT users ...