OpenAI says its models found up to 14% PPA gains on AMD flagship GPU IP OpenAI said on November 18th that its models identified up to 14% improvement in power, performance and area on AMD flagship GPU IP, measured against baselines OpenAI described as heavily optimized by AMD engineers over years of work. OpenAI said the work took less than five months and represented roughly a year of normal progress, and that the 14% figure is a best case rather than an average across the design. OpenAI also reported silicon-proven savings on its Jalapeno inference chip and a Synopsys deal to sell the workflow to other designers. OpenAI says its models found up to 14% PPA gains on AMD flagship GPU IP OpenAI reported the AMD gains on November 18th, alongside silicon-proven savings on its Jalapeno inference chip and a Synopsys deal to sell the workflow to other designers. By Ryan Merket https://runtimewire.com/author/ryan-merket · Published OpenAI said on November 18th that its models identified up to 14% improvement in power, performance and area on AMD flagship GPU IP, against baselines the company described as heavily optimized by AMD engineers over years of work. The claim was published in a post on X and detailed in OpenAI's research post. OpenAI, co-founded by president Greg Brockman and led by CEO Sam Altman, said the work took less than five months and represented roughly a year of normal progress. The 14% figure is a best case, not an average across the design. How the AMD work was done…