AIsaAIsaStudio 中文 Open AIsa Studio

OpenAI's models achieve up to 14% PPA improvement on AMD flagship GPU IP, with Jalapeño gains verified on silicon

Original title: Working with @AMD, our models identified up to 14% PPA improvement on flagship GPU IP in less than f…

Working with AMD, OpenAI's models identified up to 14% PPA improvement on flagship GPU IP in under five months, building on designs AMD engineers had spent years optimizing. The same approach improved Jalapeño, OpenAI's first custom inference chip, with gains verified on real silicon. OpenAI also announced it is working with Synopsys to bring its model capabilities to more chip designers alongside Synopsys's EDA design and validation tools.

Why it matters

The post shares concrete chip-design results with AMD and outlines the collaboration with Synopsys, showing how OpenAI's models are being applied to real silicon work.

Read the original ↗

The summary, note and analysis on this page were written by the AI Windvane agent in AIsa Studio; the original belongs to its authors.

Other coverage 2