Imagine a brain for a car that is incredibly fast but uses very little power. Instead of relying on expensive chips from overseas, this technology lets European companies design their own custom AI hardware. It's like moving from buying a pre-made suit to having a master tailor create one that fits perfectly for a specific job.
European automotive and industrial companies are overly dependent on US and Asian semiconductor suppliers, leading to supply chain vulnerabilities and high costs for AI hardware.
A 5nm reference chip platform and a lab-based prototype demonstrator system capable of running neural network applications.
If you are a vehicle OEM dealing with high costs and dependency on US or Chinese chip suppliers — this project developed a scalable 5nm processor architecture that can reduce costs by up to 90%. It allows for the creation of custom AI SoCs for advanced driver assistance systems.
If you are a chip design house dealing with the need for high-performance computing in small spaces — this project developed a unified processing platform with hundreds of cores on a single 5nm die. This provides superior power efficiency and compactness for AI applications.
If you are a smart sensor manufacturer dealing with limited power budgets for edge AI — this project developed a 100% programmable platform that handles signal processing with superior performance and area efficiency. This enables smarter, smaller sensors for industrial use.
Based on available project data, the use of this custom architecture can potentially reduce costs by up to 90% for European OEMs.
Yes, the underlying processor architecture developed by videantis is already in use in over 20 million cars to date.
Based on available project data, the project provides a reference chip platform that enables OEMs to realize their own custom AI SoCs, though specific licensing terms are not detailed.
The project is designed to meet the goals of the European Chips Act by increasing strategic autonomy and reducing dependence on Asian and US supplies.
The architecture is unified and 100% programmable, allowing it to scale from simple smart sensors up to central high-performance computing systems.
The project is led by a single German SME, videantis GmbH, which acts as the sole partner. Despite the small team size of 20 specialists, the company demonstrates significant industrial weight, having already deployed its architecture in over 20 million vehicles.
Contact videantis GmbH in Hannover, Germany
Contact us to explore licensing opportunities for 5nm AI processor architectures.
