Russia has developed the first domestically produced neural network accelerator: a breakthrough in autonomous AI
Russian engineering has made a qualitative leap in the field of hardware for artificial intelligence. A resident of the "Technopolis Moscow" Special Economic Zone has presented the country's first neural network accelerator for embedded systems — the LinQ HX module. The key feature of the development is its complete independence from foreign components and licenses. The device is based on a domestic microprocessor from the official register of the Ministry of Industry and Trade, guaranteeing technological sovereignty.
Technical Specifications and Autonomy
The LinQ HX is designed for local data processing without the need for an internet connection or cloud servers. This not only increases computing speed but also fundamentally addresses issues of cybersecurity and confidentiality. The module's main parameters are impressive:
| Specification | Value |
| Performance | 30 trillion operations per second |
| Simultaneously running neural networks | Up to 5 |
| Latency | From 1.5 to 2.3 milliseconds |
| Maximum power consumption | 40 W |
Such low power consumption combined with high performance opens doors for use in a wide variety of fields — from industrial automation to portable medical devices. The module supports a wide range of interfaces, including various sensors, video cameras, and complex control systems.
Potential Application Areas
Although the device is not currently used in urban infrastructure, its potential is enormous. Key areas include:
- Intelligent video surveillance: automatic detection of abnormal situations in real-time without sending data to a server.
- Autonomous robotics: creation of fully independent robots capable of making decisions onboard.
- Medical diagnostics: AI systems for precise image analysis and patient condition monitoring.
- Industrial Internet of Things (IIoT): predictive analytics and on-site equipment management.
The developer, the company "HiTech," obtained resident status in the SEZ in 2025. Since then, nearly 800 million rubles have been invested in the project. This is a serious bid for leadership in the edge-AI solutions segment.
My expert conclusion: The emergence of the LinQ HX is not just a local success but a strategic step for the entire industry. In a world where data security and low latency are becoming critical factors, a fully domestic solution with such characteristics could find demand not only within the country but also in the markets of friendly nations. Its application is particularly promising in systems requiring edge computing, which is one of the main trends in modern AI.