Quickly Quantum

IBM's Nighthawk r3 roadmap pairs a planned larger processor with a two-chip scaling path, while today's other stories cover cryptographic preparation, a Stacks demonstration, quantum-company finances and hardware development. We also explain post-quantum cryptography in plain language.
Source links
Quickly Quantum is an AI-voiced podcast, built and run by a real person. Nothing in this episode is financial advice.
Follow Quickly Quantum: subscribe on YouTube · Apple Podcasts · Spotify
More from Brian Lampert: Concrete Compute, the daily AI infrastructure briefing, and Space Stakes, the business of the new space race. Transcripts and every episode: quickly-quantum.kngoworld.chatgpt.site.

What is Quickly Quantum?

The daily quantum computing briefing you don't need a physics degree to follow. Every day, Quickly Quantum cuts through the hype to bring you the breakthroughs, funding rounds, policy moves, and research that actually matter, with plain-English analysis a smart non-physicist can follow. Hosted by Brian Lampert. AI-voiced, human-built, always skeptical of press releases. Nothing on this show is financial advice.

Also from Brian Lampert: Concrete Compute and Space Stakes.

IBM’s planned Nighthawk r3 targets 520 programmable qubits and 10,000 quantum operations, with a plan to link two chips into a 1,040-qubit system in 2027, according to Quantum Computing Report. That’s a serious scaling ambition, and it puts a practical question on the table for anyone building or buying quantum computing: can smaller chips work together well enough to make the larger system useful? The linked system is a plan, while the new processor itself is scheduled for 2027. Today we’ll separate IBM’s hardware targets from what its existing machines and software have already done. Welcome back to Quickly Quantum, your daily brief on the quantum frontier. It's Thursday, November 12, 2026. Let's get into it.

IBM’s move is to make growth in quantum computing a modular engineering problem: improve a processor, then connect processors so the system can grow without relying on one enormous chip. Quantum Computing Report says the Nighthawk r3 architecture is scheduled for 2027 and engineered to execute 10,000 quantum operations across 520 programmable qubits. IBM plans to link two chips via high-fidelity couplers into a 1,040-qubit system. Those are planned specifications, not a machine the article says is already operating.

A qubit is the basic unit of information in a quantum computer. Here, “programmable” means qubits available for running quantum operations; the same specification table distinguishes that number from the total physical qubits in the processor. For r3, it lists 520 programmable qubits and 2,034 physical qubits, with 1,040 qubits via a two-chip link. That distinction matters: a headline qubit count can describe different things, and the useful question is what the system can reliably do with its qubits.

The comparison IBM offers is its deployed Nighthawk r2. Quantum Computing Report lists r2 at 120 programmable qubits, 458 physical qubits and more than 7,500 accurate two-qubit operations. Two-qubit operations are the joint actions that let a pair of qubits interact; they’re essential to building a computation across a processor. The r3 target is 10,000 accurate two-qubit operations, and the article says the design uses a high-speed square-lattice architecture with rapid on-chip qubit reset. A lattice is an arrangement of qubits and connections; reset clears a qubit’s state so it can be used again.

There’s already a substantial operational record behind IBM’s next target. Since deploying in September 2026, Nighthawk r2 processors have executed more than 34 billion circuits for partners including E.ON, Algorithmiq and Q-CTRL, Quantum Computing Report says. A circuit is the sequence of operations that makes a quantum computer carry out a calculation. The same account says r2 delivered speedups of up to 19× over previous IBM Heron systems. In materials simulation benchmarks with Algorithmiq, r2’s square-lattice connectivity executed circuits 8× faster than Heron, while requiring an estimated 240× more classical memory for competing tensor-network approximations.

That last comparison gives the result useful texture: the quantum device’s performance is being judged alongside the classical computing resources needed to imitate it. A tensor network is a way classical computers represent and calculate with complex quantum states; the reported memory estimate describes the resources competing approximations would require. For you, the concrete point is that the r2 case rests on a mix of partner activity and benchmark results, while r3 remains a design target.

The modular idea depends on connections between chips as much as on the chips themselves. Quantum Computing Report says IBM demonstrated four chip-to-chip l-couplers linking Nighthawk processors, with up to 99.3% quantum state transfer fidelity. Fidelity describes how closely a transferred quantum state matches the intended state. The article says r3 will use 10 chip-to-chip l-couplers for multi-processor entanglement—linking processors so their quantum states can be connected. That’s the hardware bridge behind the 1,040-qubit plan.

But a demonstrated link is only one part of the system IBM is planning. Quantum Computing Report says IBM also demonstrated a dual-module cryogenic system cooled below 15 millikelvin in under five days. Cryogenic equipment keeps superconducting circuits at extremely low temperatures so they can operate. The account also describes work on qLDPC memory, a type of quantum error-correcting memory, where denser c-couplers reduced required Josephson junctions per module from roughly 5,000 to 2,000. Josephson junctions are circuit elements used in superconducting quantum hardware.

That’s a long way from a finished, fault-tolerant computer. The article describes Nighthawk r3 as a transitional stepping stone toward IBM Quantum Starling, targeted for 2029. A fault-tolerant system is designed to keep useful calculations going despite errors in its underlying quantum hardware. The r3 link has to work as part of a larger stack: qubits, couplers, cooling and error correction all have to cooperate. For people building these machines, that is the engineering burden hidden inside a larger qubit count.

IBM’s software announcement is aimed at making hybrid jobs easier to run. A hybrid workflow combines quantum processing with classical computing, the familiar kind used for data handling and other parts of a calculation. Quantum Computing Report says IBM introduced Managed Workflows within the IBM Quantum Functions stack, formerly Qiskit Functions, under a unified Resource Units billing model. The package provides persistent storage, GPU-accelerated classical compute and parallel orchestration for multi-billion-sample hybrid jobs. Orchestration means coordinating the different tasks and computing resources in a workflow.

The account gives two examples. Using automated sample-based quantum diagonalization pipelines, IBM and RIKEN accelerated large-scale electronic structure calculations 5× while running 100× more computations than their foundational 77-qubit study. Electronic structure calculations model how electrons behave in molecules and materials. Separately, Cleveland Clinic and RIKEN automatically processed 608 million samples for 12,000-atom biomolecular complexes 33× faster than unautomated workflows. The reported comparison here is to those unautomated workflows, not a claim that a quantum processor alone produced the whole result.

IBM also described tools for low-level control and developer work: a directed execution model powered by Samplomatic, Executor and client-side primitives; modular qiskit-mitigation and qiskit-noise-learning libraries; and pulse-level dynamic circuit tools. The article says those tools can execute 1,000,000-round repetition codes end to end alongside Qiskit Paulice spacetime codes. Repetition codes and spacetime codes are approaches to detecting or correcting errors by using patterns of quantum operations and measurements. The release also includes the Rust-based Quantum Resource Management Interface for high-performance-computing schedulers, Qiskit Fermions and alpha Qiskit Bosons utility mappers, the ffsim fermionic simulator, circuit usage profilers, and open-source Qiskit Agent Skills for AI coding assistants.

I think IBM is making a credible case that scaling is more than adding qubits: the r2 operational figures, the coupler demonstration and the hybrid-workflow examples give the roadmap substance. I’m less persuaded by the largest system figure on its own, because linking chips only matters if the connection quality and error-correction overhead let the combined machine do useful work. The article reports a transfer-fidelity demonstration, but it doesn’t establish that the planned two-chip system has already run as one processor. That distinction is the heart of the story.

I also think the software work deserves attention alongside the hardware roadmap. A quantum device has to fit into a workflow people can actually run, and the reported examples give a more tangible account of that effort than the planned qubit total alone. Still, the examples don’t settle how broadly those results generalize. IBM’s own roadmap points toward 2029 for Starling, so today’s announcements place r3 along a path rather than at the destination.

Time for the Hype Check. I’d give this a 7 out of 10 on substance: the strongest reason to score lower is that the 520-qubit processor and 1,040-qubit linked system are planned, but r2’s reported circuit activity, the demonstrated couplers and the concrete workflow examples earn a solid score.

The 2027 chip link is the hinge in IBM’s modular plan; if it works with the needed quality, IBM’s hardware and software teams gain a route toward larger systems, while anyone treating the 1,040-qubit target as a delivered machine is getting ahead of the evidence.

Before we get to the security headlines, here’s the phrase behind several of them: post-quantum cryptography. NIST explains that encryption protects confidential electronic information, including email, medical records and financial statements, from unauthorized viewers. Post-quantum cryptography means new encryption methods designed to withstand attacks from conventional computers and future quantum computers.

The reason for preparing is that some encryption relies on the difficulty conventional computers have factoring large numbers. Factoring means working out which smaller numbers were multiplied together to produce a larger one. NIST says sufficiently powerful quantum computers would not face the same difficulty with that problem. But the field still has major technical hurdles, and how formidable future quantum computers can become remains an open question.

The point of post-quantum cryptography is to base encryption on mathematical problems that are difficult for both conventional and quantum computers. NIST says its project released the first three finalized standards in 2024, and directs technical readers to its standards page and people planning a transition to its migration resources. The aim is to prepare electronic defenses before a capable machine exists, while the computing field works through the hardware challenges. This helps with one of the headlines coming up.

That security question has a specific Bitcoin angle today. Hokanews.com reports, citing Glassnode data cited by Coin Bureau, that 7.12 million BTC—35.4% of Bitcoin’s supply—are held in addresses with public keys already visible on the blockchain; the associated value is reported as over $500 billion.

The mechanism is public-key visibility. According to Hokanews.com’s account of Glassnode’s assessment, most addresses don’t reveal their associated public keys until a transaction is made from them; after that, remaining coins and later coins sent to the same address count as exposed under that assessment. The article says moving affected coins to a fresh address ends this particular exposure, while also noting that this does not eliminate every potential quantum risk. That’s a description of the article’s claim, not advice about anyone’s funds.

So the figure describes a future-facing exposure category, not coins that a quantum computer can take today. That distinction matters as Ethereum debates its own migration timetable.

Ethereum is now the next stop in the same security discussion. The Block reports that Ethlabs proposed moving Ethereum’s post-quantum upgrade to 2027, arguing that AI is accelerating quantum timelines. The group calls the plan “pqai27”; it’s a proposal, and it would need core-developer consensus.

Ethlabs also wants Ethereum to be “post-AI,” which it defines as cryptography based on as few assumptions as possible and clients hardened against frontier models that find bugs at scale, using formal verification and client diversity. The proposal distinguishes “pqai-ready,” with post-quantum consensus and an account route while the network continues at scaled capacity, from “pqai-scaled,” with full-capacity operation and post-quantum transactions and data availability.

Ethlabs says it will withdraw quick slots from Hegotá and focus those resources on pqai27. Quick slots are its proposal to make block production faster. The group wrote, We don't take this decision lightly. We pushed for quick slots because our survey showed that the community clearly wants them, and we are ready to defend this decision to the people who trusted us to deliver them. That’s a real proposed trade-off in the upgrade plan, but the 2027 date is not an approved network schedule.

From a proposed network-wide timeline to a narrower on-chain demonstration: Crypto Briefing reported that Stacks Labs completed an sBTC vault withdrawal using Lamport signatures, a type of hash-based digital signature, to show that Clarity can check those signatures. Clarity is Stacks’ smart-contract programming language; sBTC is a token designed to track Bitcoin at a 1:1 ratio.

The article calls this a proof of concept, not a network-wide upgrade. Crypto Briefing says the approach’s signature scheme, its interaction with the sBTC signer set and access for ordinary users still need details. The signer set is the group that oversees deposits and withdrawals for the peg.

The reactions underline that boundary. @DefiantNews posted on X: “STACKS TESTS POST-QUANTUM SECURITY FOR sBTC VAULT WITHDRAWALS A mainnet transfer used a one-time, hash-based signature, Stacks Labs said.” The same post says the test does not upgrade existing sBTC wallets or the token’s Bitcoin backing. That’s the important limit: a working signature check in one vault path doesn’t change the security of the underlying Bitcoin system.

Meanwhile, the quantum business picture comes with a different kind of measurement: revenue. The Quantum Insider reports that Xanadu posted third-quarter revenue of $1.3 million, up 32% from a year earlier, and ended September with $286.5 million in cash and cash equivalents. It says revenue came primarily from the Defense Advanced Research Projects Agency’s Quantum Benchmarking Initiative and partnership engagements, partly offset by completed prior-year contracts.

The company also announced a definitive agreement for approximately $140 million in Canadian federal funding, structured as a loan, and provisional approval for up to $10 million in additional DARPA project funding. The Quantum Insider reports a quarterly net loss of $32.2 million, compared with $20.3 million a year earlier, and research and development expenses increased 67% to $25.6 million.

Xanadu’s photonic approach uses light to process quantum information, so reducing optical loss—the light lost as it moves through components and connections—is central to its machine plans. The company said a new optical switch had loss of 7 millidecibels and that it believes this sets a performance record for the component. I think the financial picture is best read as company-building supported by contracts and funding, alongside substantial reported research spending.

And from commercial quantum development to defense work, proven news breaker @Qtumanalysis posted on X that Anduril is hiring a Quantum Software Engineer in Washington, DC, inside Anduril Labs. The post says the role would build quantum algorithms for “real-world defense and national security challenges” and integrate them into “existing and new defense systems.” Proven news breaker @Qtumanalysis posted on X that the work covers quantum-inspired, quantum annealing and gate-based quantum computers, and that applicants must be eligible for a Top Secret/SCI clearance.

The same post says Anduril has done this once in public, describing a January test of quantum-hybrid missile defense planning with D-Wave and Davidson Technologies, and an agreement with IonQ this summer. Those are claims from @Qtumanalysis on X; the post is the source for the hiring details. What’s notable is the stated aim to bring the work in-house, which puts quantum software inside a defense-company job description rather than leaving it only in research announcements.

On the hardware-funding side, The Quantum Insider reports that Chinese startup Zhengyuan Quantum completed an angel round to develop a planned 1,000-qubit superconducting quantum computer. The amount was undisclosed, and no delivery timeline was offered. The funding is intended to support the machine and testing of three-dimensional packaging, low-temperature connections, and integrated measurement and control technologies.

The report says Dealroom described component orders and discussions with potential customers for complete systems; it did not describe signed system orders. The technical work matters because a larger superconducting machine needs ways to route signals, manage low-temperature connections and control and measure its qubits. The funding backs development and testing of those technologies, while the planned computer remains a plan.

If you want these quantum headlines in your feed, follow Quickly Quantum wherever you listen. This has been Quickly Quantum, an AI-voiced podcast, created and built by a real human using today's cutting-edge technology. Nothing you heard on this show is financial advice. I'm Brian Lampert, and I'll catch you all tomorrow — take care!

Watching on YouTube? Hit subscribe, and tomorrow's briefing will be waiting for you. I also host Concrete Compute: a daily briefing on the AI buildout. The datacenters, the megawatts, and who actually pays for them. Find it wherever you get your podcasts.