Siemens, NVIDIA Launch Self-Verifying AI Agents for Chip Design
The expanded partnership anchors AI reasoning in deterministic physics-based EDA engines to tackle the semiconductor verification crisis.
Siemens and NVIDIA expanded their strategic partnership at DAC 2026 to deploy self-verifying agentic AI workflows for semiconductor and PCB design, moving beyond automation to continuously validated engineering outcomes.
The collaboration integrates NVIDIA's AI infrastructure—NeMo Gym, OpenShell secure runtime, and Nemotron models—into Siemens' Fuse EDA AI Agent system. Agents reason across billions of test scenarios and verify every decision against deterministic physics-based EDA engines.
The Verification Crisis
Verification consumes up to 70% of semiconductor design time. As AI chips, chiplets, and 3D ICs grow more complex, traditional verification methodologies have become insufficient.
"We're at an inflection point where the complexity of AI chips, chiplets, and 3D ICs has outpaced traditional verification methodologies," said Abhi Kolpekwar, senior vice president and general manager of Digital Verification Technologies at Siemens EDA. "Agentic AI is the natural path forward that scales with this complexity."
Measured Performance Gains
Agentic AI workflows in the Solido Characterization Suite have reduced characterization turnaround times by more than 10X while cutting token costs by 5X to 10X, according to Siemens.
The solution spans the full EDA stack: Catapult for high-level synthesis, Questa One and Veloce for verification, Solido for custom IC design, Aprisa for physical implementation, Calibre for signoff verification, and Tessent for design-for-test.
NVIDIA's Nemotron 3 Ultra reasoning model powers digital verification acceleration, leading among open models in agentic RTL benchmarking with the ACE-RTL agent.
Anchoring AI in Physics
The partnership anchors AI reasoning to proven engineering tools rather than allowing unconstrained generation. This reduces the risk of AI hallucinations in chip design—critical when errors can cost millions in fabrication runs.
"By enabling autonomous and long-running EDA AI agents to continuously validate their decisions against proven engineering tools accurately and efficiently, we help customers accelerate development, improve design quality and increase confidence in trusted engineering outcomes," said Amit Gupta, senior vice president and chief AI strategy officer at Siemens EDA.