Synopsys and NVIDIA Accelerate Chip Design Up to 30x with Grace Blackwell Platform and AI-Powered Solutions
Synopsys Inc. (Nasdaq: SNPS) has announced the next phase of its collaboration with NVIDIA, leveraging the NVIDIA Grace Blackwell platform to accelerate chip design workflows by up to 30x. This groundbreaking partnership, unveiled at NVIDIA’s GTC global AI conference, highlights Synopsys’ commitment to revolutionizing semiconductor development through advanced computing architectures and generative AI technologies.
Accelerating Semiconductor Design with NVIDIA CUDA-X and Grace Blackwell
At the heart of this collaboration is Synopsys’ integration of NVIDIA CUDA-X libraries to optimize its solutions for next-generation semiconductor development. By expanding support for the NVIDIA Grace CPU architecture and enabling more than 15 Synopsys solutions in 2025, the company aims to deliver unparalleled performance gains across a wide range of computationally intensive chip design workflows.
Sassine Ghazi, President and CEO of Synopsys, emphasized the significance of this collaboration:
“At GTC, we’re unveiling the latest performance results observed across our leading portfolio when optimizing Synopsys solutions for the NVIDIA Blackwell platform to accelerate computationally-intensive chip design workflows. Synopsys technology is mission-critical to the productivity and capabilities of engineering teams from silicon to systems. By harnessing the performance of NVIDIA accelerated computing, we can help customers unlock new breakthroughs and deliver innovation even faster.”
Jensen Huang, Founder and CEO of NVIDIA, echoed this sentiment:
“Chip design is one of the most complex engineering challenges in human history. With NVIDIA Blackwell and CUDA-X, Synopsys is cutting simulation times from days to hours—advancing chip design to power the AI revolution.”
Key Workflow Accelerations Across Synopsys Solutions
The partnership between Synopsys and NVIDIA is driving significant runtime improvements across multiple critical workflows, including circuit simulation, computational lithography, TCAD simulation, and materials engineering. Here’s how these advancements are transforming chip design:
1. Circuit Simulation: 30x Speedup with Grace Blackwell
Synopsys’ PrimeSim™ SPICE simulation workloads are projected to achieve a 30x speedup utilizing the NVIDIA Grace Blackwell platform. Today, customers already benefit from a 15x speedup using NVIDIA GH200 Superchips. These accelerated simulations enable engineers to achieve signoff with SPICE-level accuracy while reducing runtimes from days to mere hours.
2. Computational Lithography: 20x Faster Simulations
Synopsys’ Proteus™, a production-proven solution for computational lithography, is optimized for NVIDIA H100 GPUs and integrated with the cuLitho library, achieving a 15x speedup in optical proximity correction (OPC). Leveraging the NVIDIA Blackwell platform, Proteus is expected to further accelerate computational lithography simulations by up to 20x, addressing challenges at leading-edge technology nodes.
3. TCAD Simulation: 10x Faster Time-to-Results
Early results from applying GPU-enabled capabilities and NVIDIA CUDA-X libraries to Synopsys’ Sentaurus™ TCAD process and device simulation solution project a 10x acceleration in time-to-results. This solution is currently under development and is expected to be available to customers later this year.
4. Materials Engineering: 100x Faster Simulations
Synopsys’ QuantumATK®, a tool for atomic-scale modeling in semiconductor and materials research, achieves up to 100x faster time-to-results when utilizing CUDA-X libraries on the NVIDIA Hopper architecture. This enables researchers to simulate and analyze a wide range of materials with unprecedented efficiency.
Generative AI for Chip Design: Doubling Productivity with NVIDIA NIM Microservices
Beyond hardware acceleration, Synopsys and NVIDIA are advancing chip design through generative AI technologies. Customers using Synopsys’ Gen AI-powered knowledge assistant, Synopsys.ai Copilot, are already experiencing an average 2x productivity improvement compared to traditional methods. The integration of NVIDIA NIM microservices is projected to deliver an additional 2x speedup, further accelerating time-to-insights and enabling faster decision-making.
Optimizing EDA Workloads with NVIDIA Grace CPUs
In addition to leveraging the NVIDIA Blackwell platform, Synopsys is optimizing its electronic design automation (EDA) workloads for the Grace CPU architecture. More than 15 Synopsys solutions are now enabled for Grace CPUs, covering critical workflows such as circuit simulation, physical verification, static timing analysis, and functional verification. Synopsys plans to expand support for the Grace CPU architecture in 2025, further enhancing performance and scalability for its customers.
Why This Collaboration Matters for the Semiconductor Industry
The semiconductor industry is at the forefront of technological innovation, powering advancements in artificial intelligence, autonomous vehicles, 5G, and beyond. However, the complexity of chip design continues to grow, requiring faster, more efficient tools to meet the demands of modern applications.
By combining Synopsys’ leadership in EDA software with NVIDIA’s accelerated computing platforms, this collaboration addresses key challenges in chip design:
- Reduced Time-to-Market: By accelerating workflows like circuit simulation and computational lithography, companies can bring products to market faster.
- Improved Accuracy: Advanced simulations ensure SPICE-level accuracy, reducing errors and improving design reliability.
- Enhanced Productivity: Generative AI tools like Synopsys.ai Copilot empower engineers to focus on higher-value tasks, boosting overall productivity.
- Scalability: Support for NVIDIA Grace CPUs and Blackwell platforms ensures that solutions scale seamlessly with increasing design complexity.
