OpenAI used its models to deliver up to 14% PPA improvement on AMD's flagship GPU IP
OpenAI says that, working with AMD, its models produced up to a 14% improvement in power, performance, and area (PPA) on AMD's flagship GPU IP in under five months, with some changes already folded into a design sent to manufacturing. OpenAI also applied the method to its Jalapeño inference chip and tested GPT-6.2 Astra on open RISC-V cores, while partnering with Synopsys on a GPT-Synopsys EDA workflow; Anthropic reports parallel AI-driven chip optimizations using Mythos and Claude.