Imagine a computer chip that works like a human brain, seeing and thinking on the same piece of silicon. Instead of using standard electricity that gets hot, it uses super-cooled materials to move information with almost zero energy loss. It's like replacing a crowded highway of slow cars with a high-speed magnetic train that never stops.
Current AI hardware (CMOS) suffers from high power consumption and optical losses when scaling. This limits the ability to combine high-resolution vision and cognitive processing on a single, energy-efficient chip.
A Python-based thermoelectric-SPICE simulation platform and proof-of-concept superconducting Joule switches and single-photon detector arrays.
If you are a surveillance provider dealing with high power consumption in remote sensing—this project developed single-photon imagers and low-dissipation neurons that enable vision and cognition on a single chip. This allows for unprecedented sensitivity and minimal energy dissipation in the field.
If you are a hardware manufacturer dealing with latency in real-time object recognition—this project developed a platform with sub-nanosecond latency. This ensures faster reaction times for safety-critical AI systems.
If you are an imaging firm dealing with signal loss in ultra-sensitive detection—this project developed superconducting nanowire single-photon detectors (SNSPDs). This provides low-loss, single-photon-level vision for high-precision diagnostics.
Based on available project data, specific pricing is not provided, but the project emphasizes compatibility with low-cost cryostats to reduce operational expenses.
The project aims for a scalable platform that is easy to fabricate, targeting a density of 3000 neurons and over 100K synapses on less than 5 mm2.
Based on available project data, the project is currently in the research and demonstration phase; specific licensing terms have not been disclosed.
The technology is designed to be compatible with quantum applications and on-chip learning architectures, utilizing a Python-based simulation package for integration testing.
The project period runs from March 1, 2023, to February 28, 2027.
The consortium consists of 7 partners across 5 countries, showing a strong academic lean with 4 universities and 2 research centers. However, the project is led by an SME (Single Quantum BV), and the 14% industry ratio suggests a focus on translating deep-tech research into a commercial product via a specialized quantum hardware lead.
Contact Single Quantum BV in the Netherlands
Contact us to explore licensing opportunities for superconducting neuromorphic hardware.
