#quantumh2

Quantum H2

Quantinuum Breaks Quantum Volume Record, Reveals 56-Qubit H2 Upgrade, Improves Chemistry Simulations, and Grows Globally

Quantinuum, a premier quantum computing startup, developed a new Quantum Volume standard, updated its flagship quantum computer, and improved natural language processing and quantum chemistry simulations. The company signed a substantial strategic deal to boost Gulf quantum technology adoption.

Quantinuum's H2 quantum computer achieved the highest Quantum Volume (QV) to date, 2²³ = 8,388,608. IBM invented the QV metric, which measures quantum system processing performance by integrating qubit count, gate integrity, connection, and error rates. Quantinuum's five-year goal to tenfold QV per year ends with this achievement. Quantinuum claims that the QV metric is still valid and ungameable during the NISQ (Noisy Intermediate-Scale Quantum) period, allowing systems to be directly compared as they develop fault-tolerant quantum computing.

Quantinuum upgraded its flagship System Model quantum H2 from 32 to 56 trapped-ion qubits with the H2-1. Not a new system, just an upgrade. Due to the greater qubit count, commercial, “fully encompassing” simulators of the H2-1 quantum processor will no longer be accessible because a mathematically realistic simulation would require the complete memory of the world's top supercomputers.

Quantinuum CEO Rajeeb Hazra stressed that qubit quality, not number, is crucial for fault tolerance and commercialisation. Microsoft Azure Quantum officials were excited to construct more logical qubits with lower error rates using the 56-qubit computer after successful virtualisation on the 32-qubit system. Microsoft and Quantinuum partnered in 2019.

Quantinuum, a quantum software startup, disclosed the first scalable, error-corrected, end-to-end computational chemical method. This milestone shows that Quantinuum's quantum computing stack on the H2 quantum computer employing the InQuanto chemistry platform may enable quantum error-corrected chemistry simulations. The work uses logical qubits and quantum phase estimation (QPE) for molecular energy calculations to obtain quantum advantage in chemistry and fault-tolerant quantum simulations.

This technique uses the H2's scalable QCCD architecture, known for its real-time QEC decoding, conditional logic, mid-circuit measurements, high-fidelity operations, and all-to-all communication. Customers can access workflow through InQuanto. Quantinuum recommends combining quantum computing operations with supercomputing and AI to gain quantum edge in chemistry.

Quantinuum has released λambeq Gen II, the latest version of their quantum natural language processing (QNLP) program. Users may now directly translate linguistic structures and meaning into quantum circuits for hardware with λambeq Gen II, expanding on the team's 2016 QNLP breakthrough. The new version has more linguistic compositional structure than DisCoCat and is built on DisCoCirc math. The key benefits of DisCoCirc in λambeq Gen II include:

Allowing sentence structures to be built into text structures for long texts.

Concise, language-neutral compositional structure representation.

Providing theoretical evidence for quantum gadget performance improvements.

Addressing quantum machine learning model trainability via compositional generalisation.

For explainable AI (XAI), compositional interpretability is needed. More than 50,000 downloads indicate a strong user base for the λambeq application.

Quantinuum formed a joint venture with Al Rabban Capital to accelerate quantum technology commercialisation in Qatar and the Gulf. This collaboration allows Qatar to invest up to $1 billion in Quantinuum's quantum technology over the next decade, co-create local applications, and expand its workforce.

This alliance aims to expand Quantinuum's commercial reach through long-term economic partnerships. American and Qatari authorities and business leaders discussed the alliance at a Doha roundtable. U.S.-Qatari ties and corporate interests including AI innovation and quantum computing's potential to power AI data centres were discussed.