In March 2025, Synopsys announced a deepening collaboration with NVIDIA to leverage NVIDIA's Grace Blackwell platform and AI technology to accelerate chip design to new heights. This collaboration has attracted a lot of attention in the field of electronic design automation (EDA) and heralds a revolutionary improvement in chip design efficiency.
At NVIDIA's Global GTC AI Conference, Sassine Ghazi, president and CEO of Synopsys, said the company is optimizing its next-generation semiconductor development solutions to take full advantage of NVIDIA's CUDA-X library and Grace CPU architecture. The move is designed to speed up compute-intensive chip design workflows by up to 30x. Ghazi emphasized that Synopsys' technology is critical to the productivity and capabilities of engineering teams from silicon to systems, and that by leveraging the performance of NVIDIA's accelerated computing, customers are able to achieve breakthrough innovations faster.
Nvidia founder and CEO Jensen Huang also noted that chip design is one of the most complex engineering challenges in human history. With NVIDIA's Blackwell and CUDA-X, Synopsys has reduced simulation time from days to hours, advancing chip design to support the AI revolution.
At the heart of the collaboration is the use of NVIDIA's accelerated computing architecture, including the GB200 Grace Blackwell superchip, to dramatically improve the runtime of EDA workloads. Specifically, in circuit simulation, Synopsys' PrimeSim™ SPICE simulation workloads are expected to be accelerated by 30x on NVIDIA's Grace Blackwell platform. Customers are already able to achieve up to 15x acceleration with NVIDIA's GH200 superchip. This acceleration reduces the simulation time of complex circuits from days to hours, ensuring signoff with SPICE-level accuracy.
Figure: Synopsys accelerates electronic design automation with NVIDIA Grace Blackwell and AI to accelerate chip design
In the area of computational lithography, Synopsys' Proteus™ provides optical proximity correction (OPC) software and reverse imaging technology (ILT) to address the challenges of leading technology nodes. By leveraging NVIDIA's technology, Synopsys is advancing the technological revolution of this compute-intensive manufacturing process. Currently, Synopsys Proteus is optimized for the NVIDIA H100 GPU and integrated with NVIDIA's cuLitho library to achieve a 15x acceleration of OPC. Leveraging the NVIDIA Blackwell platform, Synopsys Proteus is expected to accelerate computational lithography simulations by up to 20x.
In addition, Synopsys has applied GPU capabilities and NVIDIA's CUDA-X libraries to the Sentaurus™ TCAD process simulation tool, and initial results show significant potential for acceleration. These developments demonstrate that the collaboration between Synopsys and NVIDIA is driving technological innovation in the EDA space and bringing new opportunities to the semiconductor industry.
The collaboration between Synopsys and NVIDIA combines their leadership in EDA and accelerated computing to dramatically improve the efficiency and speed of chip design. This collaboration will not only help reduce time-to-market, but will also advance the development of emerging technologies such as artificial intelligence. As technology continues to advance, it is expected that more innovations will emerge in the future to further promote the vigorous development of the semiconductor industry.
