Nvidia announced a $2 billion, multiyear investment in Synopsys stock on Monday as part of a strategic partnership to improve design and engineering across aerospace, auto, industrial and other verticals.
The partnership promises to enhance how sensors and other components work together at scale and with greater development speed inside EDA tools used in futuristic AI applications, especially for autonomous vehicles and robotics, including humanoids used in industry and medical settings.
Such AI applications are top of mind for many development teams and builders today. The partnership will likely further cement Nvidia’s role as a leading investor in future AI projects because of its leading market position in GPUs and its CUDA software.
Nvidia CEO Jensen Huang said in a press release that the move will help empower engineers to invent new products that shape the future. The principal way the partnership intends to make this transformation possible is through Nvidia’s CUDA-X libraries and GPU-accelerated computing, Huang said, to help enable simulation at faster speeds and create fully functional digital twins inside a computer.
Synopsys CEO Sassine Ghazi said the complexity of building next-gen system “demands engineering solutions with a deeper integration of electronics and physics…No two companies are better positioned to deliver AI-powered, holistic system design solutions than Synopsys and Nvidia.”
The companies will integrate Synopsys AgentEngineer technology with the complete Nvidia agentic tech stack, including Nvidia NIM microservices, Nvidia NeMo Agent Toolkit software and Nvidia Nemotron models. This integration will enable autonomous design capabilities for electronic design automation (EDA) and simulation and analysis workflows.
Nvidia and Synopsys also plan to collaborate on virtual design and testing through accurate digital twins for the semiconductor, robotics, aerospace, auto, energy, industry, and healthcare domains relying on Nvidia Omniverse and Cosmos, Nvidia said.
The partnership is not exclusive and the two companies will continue to partner with others in the EDA and design ecosystem. Both companies have agreed to make joint go-to-market approaches to reach engineering teams through the Synopsys network of thousands of direct sellers and channel partners.
Analyst Jack Gold of J. Gold Associates said the Nvidia partnership with Synopsys makes “lots of sense” because Nvidia wants a much bigger role in physical AI with its own chips and CUDA and other software.
“Having a direct link to Nvidia design assets makes sensor companies, who increasingly need to add more AI capability, more likely to deploy Nvidia hardware assets into their designs,” he said. “This is a move by Nvidia to make its smaller GPU and AI chips more attractive to sensors and other physical asset designers. It will also strengthen their CUDA monopoly should more physical sensor companies support that CUDA platform.”
What Huang and Ghazi said just after the announcement
Nvidia’s financial commitment of $2 billion to Synopsys was completed through a purchase of Synopsys shares at $414.79 per share, for about 4.8 million shares. That price was not quite 1% below the stock’s closing price of $417.01 on Friday. Synopsys shares quickly rose to more than $435 mid-day Monday, up by 4% over Friday’s close. Nvidia was up by 1.5% on Monday. Cadence, a big Synopsys competitor, dropped by 0.24% on the news.
Ghazi said the investment by Nvidia could have immediate effect in 2026 with expansion of tools used by developers. Already, some companies are using new Synopsys products, he said, without naming them.
Huang said it won’t take 10 years for the acceleration of computing needed by developers and the process could realistically last two to three years. “Engineering teams anywhere in the world of any size will have the benefit” of the partnership, Huang said. Some engineering groups will even consider acceleration immediately, he said optimistically.
Huang called the tech design and development industry one of the largest compute-intense industries not already addressed by generative AI and acceleration chips. “We’re addressing every industry where engineers create new products and factories,” he added. “This expands our use of chips to nearly every industry…It’s a huge expansion of market opportunities for both of us. It’s going to revolutionize the entire space.”
While generative AI works with text, creating software that accelerates physics and physically-based and accurate products and systems of systems “requires a whole new level of computation, an intersection of computing and the physical world...akin to robotics and that field of AI, and that’s quite new.”
Huang said current GPU adoption in the engineering industry is “quite low” even though some early tools by Synopsys have shown the value in reduced development time. Some new EDA and SDA tools already allow a process to be reduced from two weeks to just hours, Ghazi added.
“Nvidia in a lot of ways is a software company that builds a lot of chips,” Huang added. With the Synopsys collaboration and Nvidia GPUs, “adoption will happen.”
Huang clarified that the non-exclusive collaboration with Synopsys will mean Nvidia still works with Cadence and Siemens for EDA and related products. In similar manner, Ghazi said Synospsys will still sell products running Arm and X86 and from AMD and others into its customer base.