Azernews.Az

By Alimat Aliyeva

ODINN, an American startup from California, has introduced a compact supercomputer for working with artificial intelligence, which is the size of a suitcase for hand luggage, Azernews reports.

The presentation took place on the eve of CES 2026. The development was named OMNIA. According to the company, the system provides power comparable to a full-fledged data center, but does not require lengthy and expensive on-site server infrastructure construction.

In recent years, artificial intelligence has been actively used in various fields, from finance and media to aerospace and pharmaceuticals. Many companies seek to create their own AI models in order to work with their data and gain competitive advantages. At the same time, the transfer of sensitive information to cloud data centers, which may be located in other cities or countries, is not suitable for everyone.

In particular, this applies to the defense sector, government agencies, as well as data on financial transactions or medical records, which must be kept privately and not transmitted through public networks. Previously, such organizations had to build local data centers from scratch.

OMNIA aims to change this approach. The supercomputer has the same processors, graphics accelerators, memory and storage as conventional data centers, but is made in a compact package. The system is completely autonomous and can be deployed in minutes, not months.

ODINN was founded in 2023. Its goal is to reduce dependence on large data centers. To do this, the developers have taken a new approach to the issues of power supply, cooling and reliability, combining everything in one device. OMNIA uses its own closed-circuit cooling system, which allows it to work quietly even when processing large amounts of data and complex AI tasks.

The system is connected to a standard power grid and network infrastructure. It can be installed in an office, hospital laboratory, or secure facility without setting up a separate server room.

For scaling, ODINN has developed the Infinity Cube solution, a modular AI cluster that combines several OMNIA systems in one package. Each module has its own cooling and computing resources, so it does not require additional external systems such as cooling units or raised floors. This allows organizations to quickly create their own AI cluster and maintain full control over the data.

Managing such an infrastructure simplifies the NeuroEdge software layer. It is compatible with the NVIDIA software ecosystem and other AI frameworks and is responsible for task allocation and optimal cluster operation. Thanks to this, institutions can focus on using artificial intelligence rather than setting up equipment.

At CES 2026, ODINN is positioning itself not as a provider of data centers or cloud services, but as an AI infrastructure company focused on institutions with high requirements for security, privacy and minimal system delays.