SiMa.ai has unveiled Palette Neat, a new open-source development environment designed to simplify and accelerate the creation of Physical AI applications. The company says the platform enables developers to reduce application development timelines from several months to just a few days by combining agentic workflows, natural language interaction and optimized software tools within a unified engineering environment.
The launch marks a significant expansion of SiMa.ai’s software-first strategy for Physical AI and is complemented by the commercial availability of its Modalix MLSoC System-on-Module (SoM) and a new PCIe companion card. Together, the software and hardware platform is designed to support demanding AI workloads across robotics, industrial automation, automotive, aerospace, defence, drones, healthcare and intelligent vision systems while delivering high performance with low power consumption.
According to the company, Palette Neat introduces an AI-assisted development workflow that enables engineers to describe application requirements using natural language instead of manually configuring low-level software components. The platform automates much of the integration, optimization and deployment process, allowing engineering teams to devote more time to application design and innovation.
Krishna Rangasayee, Founder and Chief Executive Officer of SiMa.ai, said the company has developed an integrated platform that combines advanced AI software with proprietary silicon to simplify Physical AI deployment. He noted that the combination of Palette Neat and the production-ready Modalix System-on-Module enables developers to design intelligent systems significantly faster while reducing dependence on conventional GPU-based architectures. The objective, he said, is to shorten development cycles from months to days—and in some applications, even hours.
Palette Neat also supports migration of existing applications by allowing developers to reuse a substantial portion of previously developed software, reducing engineering effort and minimizing the cost of transitioning to new hardware platforms. The accompanying Modalix SoM has been engineered specifically for Physical AI applications and is capable of simultaneously running multiple large language models alongside computer vision and sensor-processing workloads while operating at less than 10 watts of power.
Designed as a pin-compatible replacement for widely used AI modules, it allows developers to upgrade computing capabilities without redesigning existing carrier boards, helping accelerate product deployment. SiMa.ai believes the integrated platform will help organizations streamline AI adoption across industries where intelligent machines require real-time decision-making, energy efficiency and rapid deployment.
Palette Neat has been released as an open-source platform, with supporting technical documentation available through the company’s developer resources. SiMa.ai has also announced a technical webinar scheduled for June 30 to demonstrate the platform’s capabilities and discuss emerging trends in scaling Physical AI applications.
Headquartered in San Jose, California, SiMa.ai specializes in Physical AI platforms supporting applications across robotics, industrial automation, automotive, smart vision, aerospace and defense, and healthcare industries. The company focuses on delivering software-centric AI solutions optimized for edge deployment and real-time performance.
