TLDR: SiMa.ai and Enclustra have partnered to launch a new Physical AI solution, combining SiMa.ai’s MLSoC™ Modalix technology with Enclustra’s System-on-Module (SoM) design. This innovative platform delivers high-performance, energy-efficient AI for demanding real-world environments, supporting generative AI inference at under 10 watts. It targets industries like manufacturing, robotics, aerospace, and defense, aiming to simplify the integration of advanced AI at the embedded edge.
SiMa.ai, a leader in machine learning system-on-chip (MLSoC) technology, and Enclustra, a specialist in System-on-Module (SoM) design, have announced a significant advancement in Physical AI with the joint development of a new, cutting-edge solution. This collaboration, detailed in a Bisinfotech report on October 23, 2025, introduces an innovative platform that merges SiMa.ai’s MLSoC™ Modalix technology with Enclustra’s extensive expertise in SoM design, creating a compact, industrial-grade AI solution tailored for superior performance in challenging real-world settings.
The newly unveiled platform is engineered to push the boundaries of AI deployment, particularly in applications where artificial intelligence directly interacts with physical elements, moving beyond traditional digital AI focused on ‘pixels’ to processing ‘atoms’. A key highlight of this breakthrough is its ability to support Generative AI inference while consuming less than 10 watts of power, striking an optimal balance between energy efficiency and computational prowess. The solution boasts 50 Tera Operations Per Second (TOPS) of edge-optimized computing power, setting a new benchmark for AI processing in embedded environments.
Designed with scalability and thermal stability in mind, the platform aims to accelerate time-to-market for businesses and enhance operational efficiency across a diverse range of applications. Leveraging Enclustra’s extensive experience in sectors such as aerospace, industrial, and medical, the joint platform guarantees exceptional signal integrity, precision, and durability, making it suitable for harsh operating conditions.
Philipp Baechtold, CEO of Enclustra, emphasized the company’s commitment, stating, “Enclustra is committed to pushing the boundaries of FPGA and SoM innovation. This launch strengthens our ability to deliver Edge AI solutions that empower industries to scale AI adoption efficiently.” Krishna Rangasayee, CEO and Founder of SiMa.ai, further elaborated on the strategic importance of this development: “The introduction of our Modalix SoM represents a critical expansion into Physical AI, addressing the diverse needs of robotics, automation, and autonomous systems. Together with Enclustra, we are removing integration barriers and making it remarkably simple for companies worldwide to upgrade from legacy systems to Modalix’s superior performance.”
Also Read:
- Arm Unveils Advanced Edge AI Platform, Empowering On-Device Intelligence and Innovation
- Axelera AI Unveils Europa AIPU: A New Standard for AI Accelerator Performance, Efficiency, and Cost-Effectiveness
The partnership signifies a deep co-engineering effort, aiming to redefine AI intelligence deployment in physical systems and unlock new opportunities for industries reliant on edge AI and embedded intelligence. The solution is supported by SiMa.ai’s ONE Platform for Physical AI, which includes the Palette and Palette Edgematic software suites, streamlining development with popular tools like Python, PyTorch, and OpenCV. Engineering samples of the MLSoC Modalix SoM are currently available through Enclustra’s Early Access Program, with customers able to place orders immediately for early delivery.
