Skip to content
InfoNewsLLM-specific

Advancing chip design with OpenAI models

Published
Record updated
View JSON

Summary

OpenAI reports that its models improved circuits for its custom inference chip Jalapeño, reducing standard-cell area by about 13% in the vector unit and 10% in the matrix unit. The company says the same approach gave AMD up to 14% PPA improvement over heavily optimized baselines on its flagship GPU IP, with AMD using some changes in a design sent for manufacturing. OpenAI is also working with Synopsys to integrate its models with electronic design automation tools.