SiMa.ai Unveils Modalix: A Leap Forward for Physical AI in Aerospace and Embedded Systems

SiMa.ai, a rising force in edge AI innovation, has officially launched its next-generation Machine Learning System-on-Chip (MLSoC), Modalix™, now in full production. The announcement marks a significant milestone in the evolution of Physical AI, a category that blends real-time perception, decision-making, and natural language interaction directly into embedded systems. For aerospace and defense stakeholders, Modalix represents a new frontier in low-power, high-performance AI integration.

Modalix: Purpose-Built for Embedded Intelligence

Modalix is designed to support advanced workloads, including large language models (LLMs), transformers, convolutional neural networks (CNNs), and generative AI, while operating under 10 watts. Its Arm-based architecture and native GenAI stack enable seamless deployment in environments where power, thermal efficiency, and reliability are paramount.

The chip supports essential interfaces such as camera, Ethernet, and PCIe, making it ideal for aerospace applications ranging from autonomous navigation and onboard diagnostics to secure communications and smart payload management. Its pin compatibility with leading GPU System-on-Modules (SoMs) and integration with SiMa.ai’s LLiMa software framework further streamline development for edge deployments.

Strategic Collaborations Accelerate Production

SiMa.ai’s rapid development of Modalix was made possible through partnerships with Synopsys and TSMC. Synopsys provided its AI-powered electronic design automation (EDA) suite and IP portfolio, enabling a successful first tapeout and bug-free A0 silicon. TSMC’s advanced N6 process technology ensured Modalix met stringent embedded deployment standards.

“Achieving a successful first tapeout of MLSoC Modalix illustrates the mission-critical role of Synopsys AI-powered design and IP to achieve complex SoC requirements,” said Ravi Subramanian, Chief Product Management Officer at Synopsys.

Sajiv Dalal, President of TSMC North America, added, “This collaboration underscores our commitment to driving energy-efficient chip innovations that are redefining the future of AI”.

Physical AI’s Aerospace Potential

While SiMa.ai’s announcement spans industries from robotics to retail, its implications for aerospace are particularly compelling. Physical AI enables real-time decision-making in constrained environments, a capability increasingly vital for unmanned aerial systems, satellite payloads, and next-gen avionics.

The aerospace sector is rapidly adopting edge AI to reduce latency, enhance autonomy, and improve mission resilience. Modalix’s low power draw and high compute density make it a strong candidate for integration into spaceborne platforms and airborne systems where traditional GPUs are impractical.

Moreover, the chip’s support for GenAI and LLMs opens doors for advanced human-machine interaction, predictive maintenance, and intelligent mission planning, all within the confines of embedded hardware.

A Platform Poised for Scale

With Modalix now shipping globally, SiMa.ai is positioning itself as a foundational player in the Physical AI ecosystem. Its software-centric approach lowers the barrier to adoption, allowing developers to deploy sophisticated AI models without overhauling existing infrastructure.

For aerospace engineers, systems integrators, and defense contractors, Modalix offers a scalable, production-ready solution to bring AI directly to the edge. As the demand for smarter, faster, and more autonomous systems grows, SiMa.ai’s platform could become a cornerstone of next-generation aerospace innovation.