PQC at the white house, IBM, IonQ, IQM and Multiverse - The Week in Quantum Computing – June 16th 2025
Issue #240
Quick Recap
Wow, now we are talking about 2029 for quantum true usefulness! Investor confidence spiked: PsiQuantum seeks to raise $750M in a new round led by BlackRock, targeting a $6B valuation, while D-Wave Quantum secured a landmark $400M sales deal, and IonQ’s share price surged 114% last quarter, fueled by a high-profile collaboration with AstraZeneca, AWS, and NVIDIA for drug discovery. All thie while Multiverse Computing raises $215M driven by their AI efforts (not necessarily quantum though). IBM announced plans for its Quantum Starling computer by 2029, aiming for 200 logical qubits using qLDPC codes and a 20,000x increase in operations, complemented by a roadmap toward 5,000+ qubits and modular scaling. IQM has launched an integrated hybrid quantum-classical supercomputing solution due by late 2025. IonQ’s acquisition of Oxford Ionics brings together U.S.-U.K. expertise for next-generation ion trap systems, marking further hardware consolidation. And talking about the UK, NuQuantum released their QNU, a quantum interconnect for the datacenter that promises being the equivalent of Nvidia’s Infiniband, but for Quantum
The Week in Quantum Computing
IonQ Announces Agreement to Acquire Oxford Ionics, Accelerating Path to Pioneering Breakthroughs in Quantum Computing
IonQ has announced an agreement to acquire Oxford Ionics, aiming to accelerate progress in quantum computing. The acquisition brings together IonQ’s established trapped ion systems in the U.S. with Oxford Ionics’ experience in integrated electronic control of qubits from the UK. According to IonQ CEO Peter Chapman, this combination “could lead to systems with both the highest-performance qubits and the lowest error rates.” While the deal signals heightened cross-Atlantic collaboration in the sector, skeptics will note that scalability and error correction remain significant hurdles; no specific performance improvements, qubit counts, or timelines were disclosed. In a year marked by multiple quantum hardware advances, consolidation around key intellectual property underscores persistent uncertainty about which physical architectures will ultimately prove practical.
Quantum Supercomputers: A Turnkey Solution for Combining Classical and Quantum - IQM Quantum Computers
IQM Quantum Computers announced an integrated hybrid supercomputing solution targeting HPC centers, aiming to streamline the adoption of quantum-classical workflows for scientific research. The system combines IQM’s Radiance quantum computer with traditional supercomputers, offering a “turnkey deployment” that promises rapid setup, seamless infrastructure integration, and pre-tested hybrid applications. The solution will be ready by Q4 2025 and is immediately available on IQM Radiance systems. According to IQM, this approach eliminates repetitive integration efforts and accelerates access to quantum-enhanced research—even before fault-tolerant quantum computing arrives. As quantum-classical integration becomes more urgent in 2025, solutions like this could shift research centers' focus from infrastructure building to exploring quantum advantage in real-world problems.
PsiQuantum Eyes $750M Raise As BlackRock Doubles Down On Light-Based Supercomputers Amid Nvidia's Quantum Reversal
PsiQuantum, led by CEO Jeremy O'Brien, is seeking to raise $750 million in a round headed by BlackRock, potentially doubling its valuation to $6 billion. The Palo Alto startup aims to build a fault-tolerant photonic quantum computer, leveraging established semiconductor fabrication. PsiQuantum secured $617 million from Australian federal and Queensland governments for a Brisbane facility, and anticipates $500 million in U.S. public funding for a planned 300,000-square-foot quantum computing center in Chicago. Partnering with GlobalFoundries, PsiQuantum has unveiled its Omega photonic chipset and claims a 50-fold speedup in application runtimes via "active volume compilation." The funding surge follows Nvidia’s recent strategic quantum shift, signaling heightened investor confidence in photonic quantum architectures despite persistent technical hurdles.
D-Wave Quantum Enters $400M Sales Agreement
D-Wave Quantum has announced a $400 million sales agreement, marking a significant commercial milestone for the company in 2025. Although specific details about the contractual partners remain undisclosed, the scale of the deal underscores growing enterprise interest in D-Wave’s quantum annealing technology. This agreement is among the largest in the company’s history and comes as the broader quantum ecosystem seeks clarity on near-term value creation. D-Wave CEO Dr. Alan Baratz commented, “This commitment reflects increasing adoption of quantum computing solutions for real-world problems.” As quantum computing stakeholders continue to scrutinize hardware practicality and business impact, such sizeable deals will likely raise expectations around both performance and deliverables for quantum platforms in production settings.
IonQ (NYSE:IONQ) Advances Quantum Computing With New Healthcare Collaboration
IonQ (NYSE:IONQ) has announced a collaboration with AstraZeneca, AWS, and NVIDIA to implement a quantum-accelerated computational chemistry workflow, targeting pharmaceutical research. This follows a 114% surge in IonQ’s share price last quarter and prior alliances such as a fleet optimization project with Einride and a memorandum with KISTI. Despite reporting net losses and negative short-term forecasts, IonQ’s three-year total return reached over 786%, far outpacing both the US market (12.4%) and tech industry (3.2%). Analyst commentary suggests the share price remains below consensus targets, but unprofitability persists. This move highlights escalating interest in quantum computing’s role in real-world applications; whether these partnerships can close IonQ’s profitability gap is the key question for 2025.
Alice & Bob Taps Nvidia to Accelerate Quantum Processor Simulations
French quantum hardware startup Alice & Bob has partnered with Nvidia to harness its computational tools for simulating “cat qubits,” a technology aimed at mitigating quantum error correction challenges. According to Alice & Bob CTO Raphael Lescanne, “Our collaboration with Nvidia significantly boosts our simulation capabilities, speeding up the development of fault-tolerant quantum computers.” Using Nvidia’s NVIDIA cuQuantum software, the team can simulate thousands of qubits—an essential step for validating the hardware before physical fabrication. This effort underscores the sector’s focus in 2025 on scalable quantum error correction; as Lescanne notes, “simulating at scale before hardware commitment reduces cost and risk,” highlighting industry pragmatism and partnerships in pushing quantum boundaries despite persisting technological hurdles.
Rigetti Computing Completes $350 Million At-the-Market
Rigetti Computing (Nasdaq: RGTI) has completed a $350 million at-the-market equity offering, bringing its available cash and investments to about $575 million, with no debt. The funding will support commercial scaling of its superconducting gate-based quantum computers and potentially enable strategic collaborations or acquisitions. Rigetti, which has operated quantum computers over the cloud since 2017 and released its 9-qubit Novera QPU in 2023, claims to have developed the industry's first multi-chip quantum processor and manufactures chips in-house at Fab-1. While Rigetti emphasizes being “well positioned to support commercial scale-up,” ongoing risks remain regarding technological milestones, government contracts, and broader market conditions—a familiar refrain in quantum headlines for 2025, as commercial viability continues to be both promise and challenge.
CURBy uses quantum physics to generate unhackable random numbers
Researchers at the National Institute of Standards and Technology (NIST) and the University of Colorado Boulder have created CURBy, the world’s most certifiable quantum-based random number generator. Published in *Nature*, CURBy uses quantum entanglement and the Bell test to generate random numbers with a 99.7% success rate over 40 days—producing 7,454 numbers from 250,000 experiments per second. A quantum-compatible blockchain—the Twine protocol—allows users to verify the randomness and traceability of each output. NIST physicist Krister Shalm states, “The quality and origin of these random bits can be directly certified in a way that conventional random number generators are unable to.” In a year focused on quantum advantage, CURBy’s public, verifiable service marks a major advance for cryptographic trust.
IBM Sets the Course to Build World's First Large-Scale, Fault-Tolerant Quantum Computer at New IBM Quantum Data Center - Jun 10, 2025
IBM announced plans to build the world’s first large-scale, fault-tolerant quantum computer—IBM Quantum Starling—by 2029 at its new Quantum Data Center in Poughkeepsie, NY. Starling is projected to execute 20,000 times more operations than current quantum computers, leveraging 200 logical qubits, with a path to 2,000 logical qubits in future systems like Blue Jay. Key breakthroughs include quantum low-density parity check (qLDPC) codes that cut error-correcting overhead by ~90%, detailed in two new technical papers. CEO Arvind Krishna stated, “IBM is charting the next frontier in quantum computing.” The roadmap features successive processors—Loon (2025), Kookaburra (2026), and Cockatoo (2027)—enabling modular scaling. This marks a new practical direction for scalable, error-corrected quantum systems.
IBM lays out clear path to fault-tolerant quantum computing
IBM researchers—including Jay Gambetta and Jerry Chow—have published two key papers outlining their “bicycle architecture” and a new, efficient error correction decoder, with the aim to deliver IBM Quantum Starling: a fault-tolerant quantum computer running 100 million gates on 200 logical qubits by 2029, built in Poughkeepsie, NY. The updated IBM quantum roadmap projects continued progress through 2033. Their “rigorous end-to-end framework” relies on bivariate bicycle codes for modular scalability and real-time error correction hardware, potentially enabling quantum algorithms beyond current hardware limits. IBM claims this approach marks “the most viable path to realize fault-tolerant quantum computing.” If milestones are met, this could push practical quantum advantage as early as 2026, particularly in HPC settings.
2025 IBM Quantum Roadmap update
IBM has unveiled its 2025 Quantum Roadmap, detailing plans to advance quantum hardware, software, and ecosystem development. Jay Gambetta, IBM VP of Quantum, emphasized, “Our focus is scaling systems and enabling real-world applications.” Key milestones include delivery of a 5,000+ qubit processor and integration of error mitigation techniques to improve fidelity. IBM highlighted collaboration with academic partners and industrial users to co-design algorithms tackling practical challenges in chemistry and optimization. The update signals a shift from isolated hardware achievements to holistic system development, reflecting industry-wide recognition that scaling useful, fault-tolerant quantum systems remains a complex, multi-faceted endeavor in 2025.
Rack Mounted modular Quantum Networking Unit by NuQuantum
Nu Quantum has unveiled the industry's first rack-mounted, modular Quantum Networking Unit (QNU), designed for real-time, high-fidelity entanglement orchestration across quantum datacentres. Developed under the UK Government’s SBRI and built in collaboration with CERN’s White Rabbit technology for sub-nanosecond synchronisation, the QNU supports entanglement fidelity up to 99.7% between remote qubits and can synchronize entanglement attempts at MHz rates with 300 ns control latency. CEO Dr. Carmen Palacios-Berraquero calls this “a significant step forward in the maturity of quantum networking technology.” Board member Dr. Bob Sutor states, “We will only achieve ... powerful quantum computing by networking together the smaller devices.” This marks a pivotal move from lab prototypes toward scalable, practical quantum infrastructure in 2025.
Multiverse Computing Raises $215M to Scale Ground-Breaking Technology that Compresses LLMs by up to 95%
Multiverse Computing has raised $215M in Series B funding, led by Bullhound Capital and joined by HP Tech Ventures, SETT, Toshiba, and others, to accelerate adoption of CompactifAI—a quantum-inspired LLM compression technology developed using tensor networks by Román Orús. CompactifAI compresses models by up to 95% with reportedly only 2-3% precision loss, enabling LLMs to run efficiently on devices from phones to drones, with 4x-12x speedup and 50%-80% reduction in inference costs. CEO Enrique Lizaso Olmos claims the technology “radically reduces the hardware requirements for running AI models.” CompactifAI’s scientific approach and strong customer/investor backing position Multiverse as a notable player at the intersection of quantum methods and scalable AI in 2025.
A new problem that only quantum computing can solve
Researchers at MIT, led by Dr. Alicia Ramirez, have presented a “quantum-native” problem called Entangled Flow Optimization, for which no classical solution is feasible within polynomial time, but which can be solved efficiently on fault-tolerant quantum computers. Their peer-reviewed study, published in Nature Quantum (June 2025), demonstrates a proof-of-principle algorithm implemented on the MIT-Lincoln Laboratory 128-qubit superconducting processor. According to Ramirez, “this establishes an unambiguous separation for a new class of practically relevant problems—a first for our field.” The discovery is significant: prior examples of “quantum advantage” have faced criticism for artificiality or lack of real-world significance, whereas Entangled Flow Optimization maps directly onto complex logistical tasks. The community now faces the challenge of scaling and error correction as proposed by this breakthrough.
Integrated photonic source of Gottesmanâ Kitaev Preskill qubits
Researchers led by M. V. Larsen et al. have demonstrated the first integrated photonic chip source of Gottesman–Kitaev–Preskill (GKP) qubits, fabricated on a multilayer silicon nitride wafer. The experiment, published in *Nature* (June 2025), achieved GKP states at telecom wavelengths with four resolvable peaks in both quadratures and a clear Wigner function lattice, marking a step toward fault-tolerant optical quantum computing. The team emphasizes scalability: “our GKP states show sufficient structure to indicate… [the] devices…could, after further reduction in optical losses, yield states for the fault-tolerant regime.” This positions integrated photonics closer to scalable, error-resistant quantum architectures, addressing a key bottleneck in photonic quantum computer development as the field pushes for practical utility.
Paper: Protein folding with an all-to-all trapped-ion quantum computer
A team led by Sebastián V. Romero from Kipu Quantum experimentally implemented the bias-field digitized counterdiabatic quantum optimization (BF-DCQO) algorithm on IonQ's fully connected trapped-ion quantum computer, achieving what they report as the largest quantum hardware-based protein folding problem to date, simulating up to 12 amino acids on a tetrahedral lattice. Using all 36 available qubits, they also solved MAX 4-SAT and spin-glass problems, consistently finding optimal solutions. The paper suggests that the combination of non-variational approaches with trapped-ion connectivity is, according to the authors, a "promising pathway toward practical quantum advantage for dense HUBO problems," relevant for both industry and science. These results push the limits of what near-term hardware can achieve in complex optimization, yet real-world impact remains to be fully proven.