QpiAI announces AI Enabled QpiAISense platform for worldwide deployment for controlling superconducting, spin and ion-trap qubits
QpiAI Quantum Team with QpiAISense™ system
Milpitas (California) [US]/Bengaluru (Karnataka) [India], August 24 (ANI/BusinessWire India): QpiAI, a leader in quantum computing and AI today announced QpiAISense platform for room temperature qubit control. QpiAISense accelerates ML library using discrete FPGA DSPs currently. But in future it would support massive ML compute capability as much as 8000 trillion operations per second per watt (8000 Tops/watt ) using Qpisemi's (https://www.qpisemi.tech) silicon photonics based AI20P001 to be integrated on QpiAISense platform. Qubit control and high-fidelity operation of Quantum computer, requires ML based technology, which QpiAI provides readily.
Further to accelerate optimization workload on the quantum computers, QpiAI will be integrating the Trion SoC onto QpiAISense platform. Trion will be gateway for hybrid Classical-Quantum compute software development and application. Current applications which will be run on QpiAISense will be running smoothly with both AI20P001 and Trion integration of QpiAISense.
- QpiAI begins deployment of QpiAISense room temperature Qubit controller platform, which has native ML acceleration capability to control various qubits types including Superconducting, Spin and Ion-traps.
- QpiAISense platform is thoroughly tested with Real world application like logistics, Finance, materials discovery, AI/ML applications via interfacing with QpiAI-Quantum software libraries
- Along with number of channels of 16, which is highest in Industry, ML acceleration is the key differentiator to tune the Qubits continuously to make sure high performance and lower operations error of the qubits
- QpiAISense Single unit box can control 16 channels. But it can be modularly stacked to control as many qubits as possible. QpiAISense platform can control 25, 50, 124, 256, 512 ,1024 and 2048 qubits
- QpiAIsense hardware platform and QpiAI-Quantum software platform have roadmap with integration of hybrid classical-Quantum compute chipset solution that QpiAI is working on to vertically integrate AI and Quantum processing to accelerate Quantum application adoption
- In near term future, QPIAIsense integrates Optimization processor (QpiAI Trion) and AI processor based on silicon photonics (QpiSemi AI20P001) that offers up to 8000 ToPs/W of ML acceleration capability for tuning and analytics on 1024+ logical qubits (millions of physical Qubits) and support AI and quantum applications
QpiAISense Product Details
- High speed Integrated control and readout electronics for quantum processors.
- Control and readout of up to 16 qubits.
- Trigger to AWG pulse & lt;40ns.
- Feedback Latency & lt;100ns.
- Capability for Quantum Error Correction and Error mitigation.
- Available for Shipping Q4 2022
QpiAISense unit block Performance parameters
Dr Nagendra Nagaraja CEO and Founder of QpiAI suggested, "This is a very important milestone for QpiAI. We got our Quantum compute platform correct and it is scalable. QpiAISense™ will be base platform on which we will deploy future hardware technologies, as well as software development. As we evolve into logical 1024 qubits (millions of physical qubits), we have all components to accelerate Quantum adoptions and enable our customers to really reap benefits of Quantum advancement. Our customers will be well supported with our technology roadmap to provide them with technology security to solve their domain problems confidently. QpiAI will be shipping QpiAISense starting September-2022 to key customers and partners globally and will provide them with all updates we will come up with related to AI and Quantum technologies in coming years."
Dr Manjunath R.V, a PhD from TU-Delft Netherlands who led QpiAI-Sense architecture and development and he is also General manager for Quantum hardware suggested, "QpiAI is the first Indian quantum computing company developing full-stack quantum computer with integrated quantum and AI solutions. QpiAISense offers a scalable integrated software-hardware solution for control and readout of superconducting and semiconducting qubits. The platform allows direct generation of microwave signals for control of transmon qubits without the need for additional mixers. The platform provides modules for pulse-level optimization, auto calibration with integrated machine learning accelerators and optimizer unit for execution of hybrid-quantum classical algorithms on quantum processors via a GUI-based development environment."
Lakshya Priyadarshi, Director of Quantum software suggested, "Quantum software developers write code at gate-level abstractions but the quantum machine executes those instructions at the pulse level. With its optimized pulse-level synthesizer, QpiAISense plays an important role in adding system-level optimizations to quantum software and enhances the quantum compiler modules. This development significantly improves our hardware-software co-design approach and accelerates the journey toward full-stack quantum computing solutions. The on-chip optimizer unit on QpiAISense board enables faster execution of hybrid quantum-classical algorithms that will be crucial in deploying near-term applications in chemistry, logistics, and financial services."
Dr Amlan Mukherjee, Senior Director of Quantum research added, "More the qubits merrier it is! Any practical quantum computer will require scalable control readout electronics for successful operation. That's why we have come up with QpiAISense, a plug and play control and readout hardware with built-in scalability."
Emerging market for QpiAISense is huge and Market price per unit of similar product will cost at least a million USD per unit to current users. QpiAI not only integrated high-end features, but also promises to lower TCO (Total Cost of Ownership) for its customers.
This story is provided by BusinessWire India. ANI will not be responsible in any way for the content of this article. (ANI/BusinessWire India)
Disclaimer: No Business Standard Journalist was involved in creation of this content
First Published: Aug 24 2022 | 11:45 PM IST