Synopsys Accelerates Engineering Innovation with Strategic Collaboration with NVIDIA

[Erkan Teskancan]

## Synopsys Accelerates Engineering Innovation with Strategic Partnership with NVIDIA

At the NVIDIA GTC 2026 event in San Jose, Synopsys is enabling R&D teams to design and verify intelligent products faster by integrating NVIDIA’s accelerated computing and AI solutions.

### Breaking Performance Barriers with GPU Acceleration

The collaboration between Synopsys and NVIDIA stands out as an alternative to traditional CPU-based engineering methods in industrial R&D. By integrating NVIDIA's accelerated computing and AI infrastructure into its leading EDA and multiphysics platforms, Synopsys eliminates performance bottlenecks and high infrastructure costs in large-scale semiconductor, aerospace, and automotive projects.

In materials science, the integration of Synopsys QuantumATK and NVIDIA cuEST has accelerated quantum chemistry operations by up to 30 times compared to open-source CPU models. Furthermore, migrating Ansys Fluent workloads to NVIDIA Blackwell GPUs accelerated Honda's computations by 34 times while reducing costs by 38%. These developments prove that accelerated computing is no longer just an optimization but a necessity for competition in high-accuracy simulation.

### Streamlining Chip Design with Cloud-Based AI Infrastructure

Comprehensive circuit verification is required for ultra-fast interfaces in AI connectivity. Synopsys leverages scalability with AWS cloud solutions supported by NVIDIA Blackwell GPUs. Firms like Astera Labs have reduced design verification times by 3.5 times using Synopsys PrimeSim software in the cloud environment. This combination eliminates the burden of local hardware management, allowing designers to focus on innovation.

### Bridging the Gap Between Simulation and the Real World with Physical AI and Digital Twins

Synopsys, with NVIDIA Isaac Sim and multiphysics platforms, is narrowing the gap between simulation and reality in autonomous systems and robotics. For example, digital twins created for two-armed robotic arms used in automotive manufacturing generate synthetic data by accurately simulating critical physical components such as fiber optic sensors and haptic feedback. This method significantly reduces the need for physical prototypes and ensures seamless transfer of AI models from virtual environments to the real world.

### Evolution Towards Agentic AI in Silicon Engineering

Synopsys, with the AgentEngineer platform developed in collaboration with NVIDIA, is automating complex chip design processes. This open, secure, and hardware-accelerated agentic AI platform, supporting NVIDIA NIM and Nemotron models, offers the industry's first L4-level agentic workflow for design and verification. This innovation transforms AI from a simple assistant to an autonomous management engine, allowing engineers to maintain control over complex silicon-system architectures.