rotative
the decoder · 2026年11月18日

openai 模型加速 amd 芯片設計 14% 提升

openai 與 amd 合作,使用模型優化晶片設計,在 5 個月內實現 14% ppa 提升,相當於一年的進展。

read in English · 简体中文

openai worked with amd to optimize chip design using its models, achieving up to 14 percent ppa improvement on amd's flagship gpu ip in less than five months. this progress equals about a year's worth of chip optimization on top of designs amd's engineers had already optimized for years. the same method also helped with jalapeño, openai's first custom inference chip. openai is also working with synopsys on gpt-synopsys, a model that uses synopsys's eda tools to make ai-assisted chip design more broadly accessible. amd has already folded some of the changes into a design sent to manufacturing.

read at source ↗

← rotative