Cadence unveils Maestro domain-trained AI models for chip design

Cadence Design Systems launched the Maestro AI Model Portfolio, a set of domain-specialized models built for its EDA tools and AI Super Agents. The company reports up to 30% better quality of results and 5x greater token efficiency versus general-purpose models. Nvidia is collaborating on agent security via its OpenShell sandbox.

• Maestro combines post-trained LLMs, reinforcement learning and Cadence EDA tools
• Up to 30% better quality of results and 5x token efficiency claimed
• Supports ChipStack, InnoStack, ViraStack and AuraStack AI Super Agents
• Deployable on Cadence cloud, public clouds or on-prem infrastructure