optical processing unit (OPU)
Search documents
Q/C Technologies Launches Optical Processing Unit (OPU) Initiative to Develop Proprietary Silicon Photonic Computing Architecture for AI Inference
Globenewswire· 2026-03-18 13:15
Core Insights - Q/C Technologies, Inc. has launched a new initiative to develop a proprietary optical processing unit (OPU) to address performance and energy limitations in AI inference infrastructure [1][2] Group 1: Optical Computing Advantages - Optical computing can perform matrix multiplication more efficiently than electronic GPUs, achieving O(n) or even O(1) propagation latency compared to O(n^3) for electronic devices [2] - Energy requirements for passive photonic components are minimal, which addresses a significant concern for current electronic GPUs [2] - Photonic computing offers advantages such as significantly faster clock speeds and bandwidth [2] Group 2: Development Plans - The company aims to overcome limitations in optical computing, including analog resolution, nonlinearity, and storage, through proprietary technology and partnerships [2] - Q/C is assembling a specialized engineering team to advance the optical chip program, with plans to introduce optical, semiconductor, and systems engineers, file foundational patent applications, and validate architectural designs [3] Group 3: Strategic Partnerships - Q/C has a licensing agreement with LightSolver, granting exclusive rights to innovative quantum-inspired laser-based processing units (LPUs) that solve complex problems at high speeds [4] - The collaboration with LightSolver is expected to bridge a disruptive computing paradigm for high-speed photonic computing and cryptocurrency infrastructure development [4]
Q/C Technologies Launches Optical Processing Unit (OPU) Initiative to Develop Proprietary Silicon Photonic Computing Architecture for AI Inference
Globenewswire· 2026-03-18 13:15
Core Insights - Q/C Technologies, Inc. has launched a new initiative to develop a proprietary optical processing unit (OPU) to address performance and energy limitations in AI inference infrastructure [1][3] - The company believes that optical computing can overcome the challenges faced by traditional electronic GPUs, particularly in matrix multiplication, energy efficiency, and speed [2][3] Group 1: Optical Computing Initiative - The initiative aims to design and prototype an OPU that can achieve O(n) or O(1) propagation latency for matrix multiplication, significantly improving upon the O(n) operations required by current electronic devices [2] - Q/C Technologies plans to tackle three key limitations of optical computing: analog resolution, nonlinearity, and storage through proprietary technology and partnerships [2] Group 2: Leadership and Development Plans - Q/C's Executive Chairman, Joshua Silverman, emphasizes the importance of AI in driving the next computing architecture inflection point and the company's commitment to developing an optical chip to enhance bandwidth, energy efficiency, and scalability [3] - The company is assembling a specialized engineering team to advance the optical computing program, with plans to file foundational patent applications and validate architectural designs [3] Group 3: Company Background - Q/C Technologies is focused on pioneering energy-efficient quantum-class, high-performance computing infrastructure, holding exclusive rights to innovative quantum-inspired laser-based processing units (LPUs) through a licensing agreement with LightSolver [4] - The company believes that LightSolver's technology can bridge a disruptive computing paradigm for high-speed photonic computing, particularly in cryptocurrency applications, unlocking unprecedented performance and sustainability [4]