NVIDIA Vera CPU Accelerates EDA Workloads and Enables Siemens AI Design
NVIDIA has begun deploying its internally designed Vera CPU—an 88‑core Olympus processor with LPDDR5X memory and second‑generation Scalable Coherency Fabric—in its Portland data center to develop next‑generation CPUs, GPUs and AI accelerators. Internal testing shows up to a 1.5× performance boost for selected electronic design automation (EDA) workloads, notably Cadence Jasper (formal verification) and Synopsys VCS (functional simulation). NVIDIA is working with Cadence and Synopsys to optimise these critical applications for the Vera architecture.
In parallel, Siemens announced an expansion of its strategic partnership with NVIDIA, adding NVIDIA AI technology to Siemens’ Fuse™ EDA AI Agent system. The enhanced agentic AI workflows are self‑verifying: they plan, execute and continuously check design tasks against deterministic, physics‑based EDA engines. Integrated into Siemens’ Intelligence Center X, the agents aim to improve result quality, reduce time‑to‑results, increase tool‑calling reliability and lower token‑costs. Siemens cites a ten‑fold reduction in cycle time for standard‑cell characterisation as an early benefit.
Both companies highlight how high‑performance CPU and AI capabilities can accelerate the most compute‑intensive stages of modern chip design, potentially speeding the overall semiconductor development pipeline.
Entities: Cadence Design Systems · Nvidia Corp. · Nvidia Corporation · Siemens AG · Synopsys Inc. · Synopsys, Inc. · Vera CPU
Claims
What the coverage asserts, and how well corroborated each claim is across sources.
- [● 3 SOURCES] NVIDIA is collaborating with Cadence and Synopsys to optimize critical EDA applications for the Vera CPU. (NVIDIA)
- [● 3 SOURCES] Siemens expanded its strategic partnership with NVIDIA to deliver self‑verifying agentic AI workflows for semiconductor and PCB design. (Siemens)
- [○ 1 SOURCE] The new Siemens capabilities build on its Fuse™ EDA AI Agent system and integrate NVIDIA AI technology. (Siemens announcement)
- [● 4 SOURCES] NVIDIA Vera CPU provides up to 1.5× performance improvement for selected EDA workloads such as Cadence Jasper and Synopsys VCS. (NVIDIA)
- [● 2 SOURCES] The enhanced Siemens Fuse EDA AI Agent system integrates NVIDIA AI technology and is incorporated into Siemens’ Intelligence Center X. (Siemens)
- [○ 1 SOURCE] The self‑verifying AI workflows aim to improve result quality, time‑to‑results, tool‑calling reliability, and token efficiency for long‑running engineering workloads. (Siemens announcement)
- [● 3 SOURCES] Siemens AI agents verify their actions against deterministic physics‑based EDA engines to improve result quality, time‑to‑results, tool‑calling reliability, and token efficiency. (Siemens)
- [● 2 SOURCES] Siemens self‑verifying AI agents can reduce design cycle times by more than tenfold in standard‑cell characterization. (Siemens)
- [● 2 SOURCES] Vera CPU contains 88 custom Olympus cores, LPDDR5X memory, and second‑generation Scalable Coherency Fabric. (NVIDIA)
- [● 2 SOURCES] Vera CPU is being deployed in NVIDIA’s Portland data center for internal chip design workloads. (NVIDIA)