Micron Technology Invests in SiMa.ai to Accelerate High Performance Physical AI at the Edge

SiMa.ai announced a strategic investment from Micron Technology today to scale production of its Physical AI solutions. This collaboration focuses on integrating Micron's high performance memory directly into the SiMa.ai compute architecture. The partnership aims to deliver superior performance per watt for robotics, autonomous vehicles, and industrial automation. The core of this effort involves the Modalix MLSoC, which now leverages Micron LPDDR5X memory. This combination allows for complex workloads like Large Language Models and Vision Language Models to run natively on edge devices. By optimizing the memory hierarchy, the platform overcomes thermal and power constraints typical of industrial hardware.

CEO Krishna Rangasayee noted that physical AI requires a new approach where memory and compute are tightly co-optimized. The flexibility of the SiMa.ai Palette software further simplifies the deployment of these complex models. This hardware synergy is designed to decrease time to market for global customers building next generation infrastructure.

Micron backing follows a period of significant growth for SiMa.ai, which has raised over $355 million to date. The company valuation recently approached $1 billion following an $85 million Series C round in late 2025. This latest injection of capital strengthens an ecosystem that already includes partners like Arm and TSMC.

The Physical AI sector is expected to reach nearly $960 billion by 2033 as demand for autonomous systems grows. Hardware solutions currently represent over half of this market total revenue. As labor shortages drive the need for automation, SiMa.ai is positioned to lead the deployment of spatially aware machines.

Developers can access these capabilities today through System on Modules that feature the integrated Micron memory. These modules provide a direct upgrade path from prototypes to full scale production environments. More details on these solutions can be found at the official SiMa.ai website.