NVIDIA has made a new splash in the race to dominate the era of artificial intelligence. The company announced a multi-year strategic alliance with Synopsys, one of the world’s leading software providers for chip design and system simulation, accompanied by an investment of $2 billion in Synopsys common stock at a price of $414.79 per share.
The declared goal is ambitious: to combine NVIDIA’s GPU-accelerated computing and AI with Synopsys’s design, verification, and simulation tools to revolutionize how intelligent products are conceived and validated, ranging from semiconductors to airplanes, cars, and industrial equipment. This move also strengthens NVIDIA’s role not just in AI hardware but also in the “tool layer” used by engineers to create the next generation of chips and systems.
From atoms to digital twins: what NVIDIA and Synopsys are aiming for
The collaboration is structured along several fronts that, taken together, tell a clear story: bringing classical design and engineering into the realm of agentic AI, massive simulation, and digital twins.
Key points of the agreement include:
- Accelerating Synopsys software with CUDA-X and “physical AI”
Synopsys will optimize its extensive catalog of applications — chip design, physical verification, electromagnetic analysis, optical simulations, molecular simulation, or multiphysics — to leverage CUDA-X libraries and NVIDIA’s physical AI technologies. The promise: shift from workflows that take days on CPUs to much faster GPU-based simulations. - Bringing agentic AI to electronic design automation (EDA)
The two companies will integrate Synopsys AgentEngineer, their agents technology for EDA and simulation, with NVIDIA’s agentic AI stack: microservices NVIDIA NIM, NeMo Agent Toolkit tools, and Nemotron models. The goal is to enable autonomous AI agents to explore design variants, run simulations, analyze results, and propose iterations with minimal human intervention. - Connecting physical and digital worlds through digital twins
The alliance will utilize NVIDIA Omniverse and NVIDIA Cosmos to create highly faithful digital twins that allow for designing, testing, and validating complex products in virtual environments: factories, robots, vehicles, medical devices, energy systems, or communications infrastructure. - Turnkey cloud solutions
Both in public cloud and hybrid environments, Synopsys tools will feature native GPU support so engineering teams of various sizes can access this compute capacity without building their own server farms.
Additionally, NVIDIA and Synopsys will coordinate joint go-to-market initiatives to bring these accelerated solutions to a broad customer base, including major chipmakers and sectors like aerospace, automotive, energy, and healthcare.
A major financial and strategic move
The $2 billion investment makes NVIDIA one of the main shareholders of Synopsys, with an estimated stake close to 2.6% of the company’s equity based on publicly available data.
The stock market response was swift: Synopsys shares rose about 7–8% following the announcement, in a context where analysts have been viewing the company as one of the big winners of the AI boom, given its key role in designing the chips that make AI possible.
For NVIDIA, the deal fits into a broader strategy of cross-investments within the AI ecosystem, which in recent months has included commitments with key hardware and software players. Several analysts already describe a “web of participations and alliances” reinforcing NVIDIA’s position not only as a GPU provider but also as a center of gravity across the entire AI stack, from models to design tools.
From traditional EDA to comprehensive simulation era
This move comes at a transformative moment for Synopsys. In July 2025, the company completed its acquisition of Ansys, a leader in multiphysics simulation software (structures, fluids, thermodynamics, electromagnetism), in a deal valued at around $35 billion, the largest in its history.
With this acquisition, Synopsys has shifted from being a giant of traditional EDA — software for chip design and verification — to also controlling one of the pillars of physical system simulation used in automotive, aerospace, energy, and telecom sectors. The partnership with NVIDIA now adds the missing piece:
- Massive GPU acceleration,
- Agentic AI applied to engineering flows, and
- Real-time 3D visualization and collaboration via Omniverse.
If successfully integrated, this could redefine how products are designed: moving away from a linear CAD → simulation → physical prototype process to a continuous cycle where AI models and digital twins explore thousands of variants before a single physical prototype exists.
What it means for “hands-on” engineering
Beyond the numbers and names, the core message is that engineering is shifting from purely calculation-focused to data and AI-driven.
Practically, the NVIDIA-Synopsys collaboration points toward:
- Much faster and cheaper simulations
Where large CPU clusters and days of computation were once necessary, GPUs and AI models now drastically reduce times, energy costs, and iteration costs. This is especially relevant in advanced semiconductor design, where each cycle can consume millions. - Automatic exploration of design spaces
AI agents can propose design variants, run simulations, analyze results, and discard options, leaving final validation and high-level decisions to engineers. This approach amplifies the “intellectual capacity” of teams without increasing headcount. - Digital twins supporting the entire lifecycle
From concept to field operation, digital twins enable testing control strategies, predictive maintenance, or configuration changes without touching the real system. For industries like automotive, robotics, energy, and healthcare, this can mean fewer failures, less product recall, and more improvement iterations. - Cloud access for medium and small teams
The “cloud-ready” aspect of the partnership aims to democratize these capabilities. Not all manufacturers can afford their own super data centers but can access GPU and specialized software on demand.
A non-exclusive partnership within a scrutinized ecosystem
Both NVIDIA and Synopsys emphasize that the alliance is not exclusive. The GPU manufacturer will continue working with other players in the EDA and simulation space, like Cadence or other industry providers, while Synopsys maintains agreements with NVIDIA competitors, including alternative CPU and GPU vendors.
However, the move intensifies the debate over the growing power of a few giants across the AI value chain: from chips to design software, to models and cloud services. Some analysts already speak of a “stack lock-in” that could make it harder for new entrants to break into the market without passing through major platforms.
For now, what’s clear is that the boundary between chip design, system simulation, and AI integration is becoming increasingly blurred. The NVIDIA–Synopsys duo aims to be the default option for those wanting to operate at this frontier.
According to Jensen Huang, combining CUDA, AI, and digital twins aims to enable simulations “from atoms to complete systems” within the computer. If the partnership lives up to its promises, much of tomorrow’s engineering could begin as a GPU-based experiment… before becoming physical hardware.
via: Press release