
SiMa.ai is a company dedicated to accelerating the practical application of physical AI in a wide range of markets, including robotics, automotive, industrial, aerospace, defense, and healthcare. We provide the industry's best physical AI platform, leading the market with superior ease of use, outstanding performance, and high power efficiency. Our experts, with extensive experience in both technology and business, provide strong support for our customers' AI adoption.
Processes more frames per watt than competing devices, delivering unparalleled efficiency
Supports multimodal AI and generative AI at the edge, enabling deployment to a wide range of applications

25mm x 25mm

MLSoC™ Modalix Block Diagram
It is pin-compatible and form-factor compatible with major GPU SoMs, enabling seamless replacement and rapid integration. Compact and power-efficient, it integrates MIPI, memory, and necessary I/O, providing the functionality needed to scale Physical AI.

System-on-Module (SoM)

PCIe HHHL
This is a complete hardware/software suite for evaluating and prototyping edge AI applications using the latest MLSoC™ Modalix SoM (System-on-Module). Designed for rapid development and deployment, it enables benchmarking and proof-of-concept (PoC) in a short space of time when combined with SiMa.ai's Palette™ software. Compact yet energy-efficient, it is ideal for industrial equipment, robotics, and vision applications, supporting real-time inference of computer vision, transformer models, and generative AI workloads.


Specializing in vision AI, it achieves low-power, high-performance processing for industrial, robotic, and medical equipment.

31mm x 31mm

MLSoC™ Davinci Block Diagram

Performance comparison with devices from major GPU vendors


PCIe board for powering the edge of data center servers and industrial/commercial PCs, achieving low power consumption of 10-15W under typical workloads

HHHL Production Board

Dual M.2 Production Board
The MLSOC™ Development Kit 2 features SiMa.ai's MLSOC™ machine learning silicon and includes everything needed to evaluate, prototype, and demonstrate standalone computer vision machine learning applications. In addition to the HHHL PCIe development board, the MLSOC™ Development Kit 2 also includes the Palette™ development tool. The HHHL PCIe development board has a variety of interfaces and supports both standalone operation and integration into PCIe platforms, making it suitable for a wide range of developers.


It runs as a Docker container on your development machine and provides all the tools you need to develop, compile, and deploy high-performance AI applications.

Accelerate your Physical AI development with SiMa.ai Edgematic on AWS Marketplace. This low-code/no-code visual platform makes it easy to build, test, and deploy AI/ML pipelines on edge devices. Perfect for both experienced and inexperienced developers.


I replaced a different company's GPU with SiMa.ai! [Overview]
Understand the ease of development and porting using SiMa, understand the preparations for evaluation, and understand the actual work procedures.

I replaced a different company's GPU with SiMa.ai! [Preparation]
Understand the environment required to run SiMa and the differences between each development environment

I replaced a different company's GPU with SiMa.ai! [Environment setup part 1]
Learn how to connect the SiMa development kit to the host machine

I replaced a different company's GPU with SiMa.ai! [Environment setup part 2]
Learn how to set up the SiMa development kit environment

I replaced a different company's GPU with SiMa.ai! [Environment setup part 3]
Understand the procedure for building the host machine environment in SiMa

I replaced a different company's GPU with SiMa.ai! [Implementation]
Learn how to build a system using the SiMa development kit
Solution Technology Department 3
03-6361-8081Reception hours: Weekdays from 9:00am to 5:00pm