The semiconductor industry is advancing at an unprecedented pace, driven by increasingly complex architectures, advanced process nodes, and rising performance demands. To help engineering teams keep up with this rapid evolution, Keysight Technologies has unveiled a new Machine Learning Toolkit within its latest Device Modeling Software Suite, designed to dramatically accelerate device modeling, parameter extraction, and Process Design Kit (PDK) development.
By applying artificial intelligence and machine learning to traditionally manual workflows, Keysight is enabling faster Design Technology Co-Optimization (DTCO) and significantly reducing time-to-market for advanced semiconductor designs.
Addressing the Growing Complexity of Semiconductor Design
Modern semiconductor development is shaped by innovations such as gate-all-around (GAA) transistors, wide-bandgap materials like GaN and SiC, and heterogeneous integration approaches including chiplets and 3D stacking. While these technologies unlock major performance and efficiency gains, they also introduce substantial modeling challenges.
Traditional compact modeling relies heavily on physics-based approaches and manual parameter extraction. Engineers often need to tune hundreds of interdependent parameters across multiple operating conditions—a process that can take weeks and still struggle to achieve optimal predictive accuracy. As design cycles compress and complexity increases, these legacy workflows are no longer sustainable.
This shift has made AI/ML-driven semiconductor modeling not just beneficial, but essential.
How Keysight’s Machine Learning Toolkit Changes the Game
Keysight’s new Machine Learning Toolkit, available within Device Modeling MBP 2026, introduces a modern framework that combines advanced neural network architectures with ML-based optimization techniques. The result is a highly automated, scalable solution that transforms how device models are created and validated.
With the toolkit, parameter extraction steps can be reduced from more than 200 manual adjustments to fewer than 10 automated steps. This breakthrough enables faster PDK delivery, accelerates DTCO workflows, and helps design teams respond more quickly to changing process technologies.
Key Features and Benefits
Accelerated Parameter Extraction
The Machine Learning Toolkit automates global optimization of more than 80 parameters in a single run. By capturing secondary effects, temperature variations, and dynamic behaviors, it improves predictive accuracy across DC, RF, and large-signal domains—while eliminating repetitive manual tuning.
Automated, Customizable Workflows
The toolkit integrates seamlessly with Keysight’s existing Device Modeling platform and supports Python-based customization, allowing teams to adapt automated flows to their specific modeling requirements.
Scalable Across Multiple Technologies
The solution is designed for reuse across a wide range of device types, including FinFET, GAA, GaN, SiC, and bipolar technologies. This scalability ensures consistent, repeatable modeling workflows across multiple process nodes.
Faster and More Efficient DTCO
By enabling tighter feedback loops between device and circuit design, the toolkit reduces PDK development cycles from weeks to days—significantly improving DTCO efficiency and accelerating time-to-market.
Industry Impact and Expert Insight
According to Nilesh Kamdar, General Manager of Keysight EDA, AI and machine learning are fundamentally reshaping compact modeling methodologies. With the Machine Learning Toolkit, Keysight empowers customers to deliver higher-quality, more predictive models in a fraction of the time, helping them stay competitive as semiconductor technologies continue to evolve.
By reducing development risk and increasing modeling accuracy, AI/ML-driven workflows allow semiconductor companies to innovate faster while maintaining confidence in their designs.
Additional Enhancements Across Keysight’s Device Modeling Portfolio
Alongside the Machine Learning Toolkit, Keysight has introduced several complementary updates:
- Device Modeling MQA 2026 adds new aging model QA rules for OMI and MOSRA.
- Device Modeling WaferPro 2025 introduces remote-control capabilities for low-frequency noise testing using A-LFNA.
- A-LFNA 2026 adds new low-frequency noise stress test functionality, enabling seamless transitions from stress testing to noise measurement.
Accelerate Your Semiconductor Innovation
As AI and machine learning continue to redefine semiconductor engineering, Keysight’s Machine Learning Toolkit represents a major step forward in automated device modeling, PDK development, and DTCO optimization.
To learn more about how Keysight can help accelerate your design workflows and improve modeling accuracy, explore Keysight Device Modeling Solutions today.
About Keysight Technologies
At Keysight (NYSE: KEYS), we inspire and empower innovators to bring world-changing technologies to life. As an S&P 500 company, we’re delivering market-leading design, emulation, and test solutions to help engineers develop and deploy faster, with less risk, throughout the entire product life cycle. We’re a global innovation partner enabling customers in communications, industrial automation, aerospace and defense, automotive, semiconductor, and general electronics markets to accelerate innovation to connect and secure the world. Learn more at Keysight Newsroom and www.keysight.com.
