Insights:
- Cadence and NVIDIA expand partnership to accelerate semiconductor and AI system design using agentic AI and digital twins.
- AgentStack orchestrates multi-agent workflows, boosting chip design productivity up to 10X and speeding workflow iterations from days to hours.
- Collaboration integrates physics-based simulation and robotics AI to close the sim-to-real gap for autonomous machines using NVIDIA Isaac and Cadence Physical AI Stack.
- AI factory digital twins optimize power efficiency and performance, enabling up to 32% more tokens per watt for hyperscale AI deployments.
- Leading customers like Samsung, Honda R&D, and SK Hynix are already benefiting from these accelerated, AI-driven engineering workflows.
At CadenceLIVE Silicon Valley 2026, Cadence announced an expanded partnership with NVIDIA aimed at transforming engineering productivity through the fusion of agentic AI, physics-based simulation, and digital twin technology. This collaboration targets faster and more efficient semiconductor design, physical AI systems, and hyperscale AI factories by integrating Cadence’s AI-driven EDA tools with NVIDIA’s CUDA-X accelerated computing and Omniverse digital twin platforms.
Central to this effort is Cadence’s AgentStack, an AI “head agent” that coordinates complex chip design workflows, delivering up to 10X productivity improvements and significantly compressing iteration cycles. Beyond chip design, the partnership addresses real-world robotics challenges by bridging the sim-to-real gap through high-fidelity multi-physics simulations combined with NVIDIA’s Isaac robotics simulation environment, enabling safer and faster deployment of physical AI systems.
Moreover, the companies are pioneering AI factory digital twins to optimize energy efficiency and maximize tokens per watt, a crucial metric for large-scale AI training and inference data centers. Early results demonstrate substantial efficiency gains and potential billions in annual revenue uplift. Industry leaders such as Samsung, Honda R&D, and SK Hynix are already leveraging these advanced solutions to accelerate innovation and reduce time to market.
Note: All images and Products & Trademarks mentioned in this article are copyrighted to companies:
Cadence, NVIDIA, Ascendence, Argonne National Laboratory, Honda R&D, Samsung, SK Hynix