Haiqu

Advanced quantum simulations should not require months of custom engineering or a fortune to run. In our blog, we show how Haiqu’s first 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗦 helps build and execute frontier quantum R&D workflows: - framing the problem, - validating baselines, - optimizing circuits and mitigating noise, - orchestrating execution Check the case studies: 1. An extended 40-qubit SKQD workflow for finding the ground state of the single-impurity Anderson model. The automatically designed hybrid algorithm used Haiqu’s circuit compression to improve QPU results and reduce the classical compute resources required for post-processing. 2. A quantum pipeline for reconstructing the magnetic excitation spectrum of CuDCl, a quasi-1D spin-½ Heisenberg material. HaiquOS enabled optimized state preparation and full-spectrum simulation on quantum hardware, with all circuits of up to 14,261 gates executed in under 5 minutes. Read the blog: https://lnkd.in/ebt_2z3B Tune in for updates.

Most quantum R&D teams aren't limited by the hardware. They're limited by the software between them and the hardware. What if your R&D team could identify the right problem faster? Design executable experiments without weeks of setup? Run them efficiently on real quantum hardware  and actually get usable results? That's exactly what we built. The first 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗢𝗽𝗲𝗿𝗮𝘁𝗶𝗻𝗴 𝗦𝘆𝘀𝘁𝗲𝗺 𝗳𝗼𝗿 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗥&𝗗 𝘁𝗲𝗮𝗺𝘀 is now available for enterprises and academic labs. After successful beta-testing with our industrial partners, 𝗛𝗮𝗶𝗾𝘂𝗢𝗦 is ready for teams who want better results, faster iteration, quicker application development and more output per budget. Our OS combines quantum research agents with Haiqu's proprietary software stack so teams can go from early idea to testable prototype without the usual friction. Orders of magnitude less money per experiment. Faster researcher onboarding. Real performance standards for quantum R&D — not theoretical ones. From the very beginning, our role in the quantum ecosystem has been clear to us. We are the toolmakers for a new creative frontier. And the tools are ready. 💡 Read the full press release here: https://lnkd.in/dFNb7-Ur 👉 Contact our team to learn more: https://lnkd.in/dNvrtSdj Richard Givhan Mykola Maksymenko

This month our CTO Mykola Maksymenko joined Scott Buchholz (Federal CTO, Deloitte), Dr. Mekena McGrew, PhD (Quantum Information Lead, Deloitte) and Steven Gibson (CCO, Strangeworks) for a Deloitte DBriefs session: "A strategic quantum path: value now, learn fast." One theme came through clearly: when companies set up a quantum program, the key is to focus on methods that build toward quantum readiness and not on classical methods that happen to carry a quantum label. The methods that do build readiness are those that plug into the quantum pipeline helping teams simulate, compile, compress, and optimize quantum algorithms so they can run on real hardware. That's what separates a quantum program from an HPC rebrand. If you missed the live session, the full recording is available on Deloitte's DBriefs platform here: 👉 https://lnkd.in/dek5zJ9a

Haiqu and HSBC Demonstrate Scalable Quantum Encoding for Financial Models We have just published joint research with HSBC in Physical Review Research — tackling one of the biggest roadblocks between quantum computing and real-world finance. To use quantum hardware for risk modeling or portfolio optimization, engineers first need to load financial data into quantum circuits. The challenge? Translating real-world probability distributions into a quantum-ready format has been slow and the number of operations grows exponentially, making it impractical on today's hardware. Our team developed a faster, more scalable method for encoding complex financial distributions into quantum circuits, cutting through the bottleneck that has held back progress in quantum finance. The approach was validated on finance-relevant models, including heavy-tailed Lévy distributions commonly used to capture extreme market events such as crashes and sudden price swings, and executed on IBM Quantum hardware using up to all 156 qubits of the device. Check our press release: https://lnkd.in/d_rtgPaj and read the paper in Physical Review Research here:https://lnkd.in/dgSB_fqz

Quantum engineers today are forced to patch together their custom workflows from multiple disconnected tools. The real pain lies in fragmentation and combinatorial incompatibility. There are multiple tools at each stage that do not always work together. This all distracts from the ultimate goal — turning an ambiguous research or business question into a real application. There could be a smooth end-to-end workflow: from application design to algorithm optimization to runtime execution. Built as a single system for quantum R&D teams working at the edge of what is possible. That’s the shift quantum R&D needs. And it’s starting to happen. Stay tuned.

As we enter the early fault-tolerant era, logical qubits won’t be “perfect.” The opportunity is a hybrid move: apply error mitigation on alongside error correction to push useful depth sooner — without waiting for massive code distances. At Haiqu, we run a weekly reading club to stay close to the research frontier. In this series, we share curated notes from our discussions: the papers, the core insight, and the practical takeaway. This week’s theme: error mitigation for logical qubits — what changes when the object you’re mitigating is already encoded. 𝗣𝗮𝗽𝗲𝗿𝘀 𝘄𝗲 𝗱𝗶𝘀𝗰𝘂𝘀𝘀𝗲𝗱 (𝗮𝗻𝗱 𝘄𝗵𝘆 𝘁𝗵𝗲𝘆 𝗺𝗮𝘁𝘁𝗲𝗿) 𝗪𝗮𝗵𝗹 𝗲𝘁 𝗮𝗹. (𝟮𝟬𝟮𝟯) — ports Zero-Noise Extrapolation (ZNE) into the FT setting by treating the code distance as the noise-scaling knob (“distance-scaled ZNE”). https://lnkd.in/ed-wZKfR 𝗭𝗵𝗮𝗻𝗴 𝗲𝘁 𝗮𝗹. (𝟮𝟬𝟮𝟱) — experimental-style demonstration of quantum error mitigation on logical qubits (ZNE applied in a QEC workflow, with practical choices about how to amplify/scale noise). https://lnkd.in/dCNrwnHn 𝗝𝗲𝗼𝗻 & 𝗖𝗮𝗶 (𝟮𝟬𝟮𝟲) — flips the stack: do linear error mitigation on physical qubits inside a logical-qubit workflow, and analyze how it composes with QEC. https://lnkd.in/drRZpi7t Follow Haiqu for more Reading Club notes like this

We asked ourselves: How close is quantum CFD actually to practical applications? Let's be honest about that. Classical methods are highly optimized. The quantum hardware is noisy. The state-preparation and readout bottleneck is real. Quantum CFD circuits are large and complex, and have often never touched a real quantum processor. So far. Last week, in collaboration with Quanscient, we announced a breakthrough algorithm for scalable computational fluid simulations on IBM Quantum computers. The breakthrough was possible thanks to the quantum middleware layer. This is what turned a theoretical experiment into a practical one. We ran a nonlinear fluid-flow benchmark with an immersed obstacle on IBM quantum hardware in a hybrid classical-quantum simulation loop. Compared to earlier quantum QLBM results, this benchmark added nonlinear flow, internal geometry, boundary-driven dynamics, coupled flow fields, and repeated hybrid time stepping. The kind of structure that real industrial problems actually depend on. For everyone who models how air, water, and other fluids behave around objects, read our new technical blog and explore how quantum can help in your organization. For all quantum researchers and engineers, check it and let us know what you think. We want to hear from both sides. Check the blog here: https://lnkd.in/dbDYwfyM

Quanscient is hosting a webinar with us next Thursday to discuss our latest paper and the implications for the industry. A well-designed quantum algorithm is one thing. Getting it to run on hardware is another. Haiqu's middleware layer closes that engineering gap by enabling on-hardware execution through compilation, transpilation, error mitigation and runtime orchestration. Valtteri Lahtinen from Quanscient and our Principal Scientist, Maciej Koch-Janusz, will walk you through our latest results, our technology and next steps towards more complex simulation of computational fluid dynamics on quantum hardware. Join us to learn how each layer of the middleware stack determines whether a quantum algorithm produces meaningful results in practice. Register here and share with your network: https://lnkd.in/dKJNVs8Z

Haiqu and Quanscient have conducted a benchmark of 15-step nonlinear fluid dynamics around an obstacle on IBM's Heron hardware. The most complex, publicly documented Quantum Lattice Boltzmann Method (QLBM) hardware demonstration to date. At the heart of this work is a novel algorithm based on the One-Step Simplified LBM (OSSLBM) combined with Haiqu's middleware for circuit optimization and error reduction. This breakthrough builds on a journey that began at Airbus&BMW Quantum Mobility Challenge in 2024, where our joint team ran the first multi-step QLBM simulations on real hardware, executing 3 steps of fluid evolution on a 64×64 grid using just 16 qubits on IonQ Aria 1 via Amazon Web Services (AWS). From 3 steps to 15. From linear to nonlinear dynamics. From a point-like disturbance to flow around an obstacle. “𝘞𝘦 𝘯𝘦𝘦𝘥 𝘮𝘰𝘳𝘦 𝘸𝘰𝘳𝘬 𝘭𝘪𝘬𝘦 𝘵𝘩𝘪𝘴 𝘵𝘰 𝘢𝘤𝘩𝘪𝘦𝘷𝘦 𝘪𝘯𝘥𝘶𝘴𝘵𝘳𝘪𝘢𝘭𝘭𝘺 𝘳𝘦𝘭𝘦𝘷𝘢𝘯𝘵 𝘲𝘶𝘢𝘯𝘵𝘶𝘮 𝘴𝘰𝘭𝘶𝘵𝘪𝘰𝘯𝘴” comments Oleksandr Kyriienko, Professor and Chair in Quantum Technologies at the The University of Sheffield.  “𝘛𝘩𝘪𝘴 𝘪𝘴 𝘢𝘯 𝘪𝘯𝘵𝘦𝘳𝘦𝘴𝘵𝘪𝘯𝘨 𝘢𝘯𝘥 𝘵𝘪𝘮𝘦𝘭𝘺 𝘤𝘰𝘯𝘵𝘳𝘪𝘣𝘶𝘵𝘪𝘰𝘯 𝘵𝘰 𝘲𝘶𝘢𝘯𝘵𝘶𝘮 𝘊𝘍𝘋. 𝘐𝘵 𝘱𝘳𝘰𝘱𝘰𝘴𝘦𝘴 𝘢 𝘮𝘰𝘳𝘦 𝘧𝘭𝘦𝘹𝘪𝘣𝘭𝘦 𝘲𝘶𝘢𝘯𝘵𝘶𝘮 𝘓𝘉𝘔 𝘧𝘳𝘢𝘮𝘦𝘸𝘰𝘳𝘬 𝘸𝘩𝘪𝘭𝘦 𝘬𝘦𝘦𝘱𝘪𝘯𝘨 𝘵𝘩𝘦 𝘤𝘰𝘳𝘦 𝘢𝘭𝘨𝘰𝘳𝘪𝘵𝘩𝘮 𝘦𝘧𝘧𝘪𝘤𝘪𝘦𝘯𝘵, 𝘢𝘯𝘥 𝘪𝘵 𝘴𝘵𝘳𝘦𝘯𝘨𝘵𝘩𝘦𝘯𝘴 𝘵𝘩𝘦 𝘤𝘢𝘴𝘦 𝘸𝘪𝘵𝘩 𝘢𝘱𝘱𝘭𝘪𝘤𝘢𝘵𝘪𝘰𝘯𝘴 𝘳𝘢𝘯𝘨𝘪𝘯𝘨 𝘧𝘳𝘰𝘮 𝘭𝘪𝘯𝘦𝘢𝘳 𝘢𝘤𝘰𝘶𝘴𝘵𝘪𝘤𝘴 𝘵𝘰 𝘐𝘉𝘔-𝘘𝘗𝘜-𝘢𝘴𝘴𝘪𝘴𝘵𝘦𝘥 𝘯𝘰𝘯𝘭𝘪𝘯𝘦𝘢𝘳 𝘧𝘭𝘰𝘸 𝘴𝘪𝘮𝘶𝘭𝘢𝘵𝘪𝘰𝘯𝘴.” Read more in the press release here: https://lnkd.in/dc96_wr8

𝐇𝐚𝐢𝐪𝐮 𝐏𝐫𝐨𝐣𝐞𝐜𝐭𝐬 𝐁𝐫𝐞𝐚𝐤𝐢𝐧𝐠 𝐑𝐒𝐀 𝐄𝐧𝐜𝐫𝐲𝐩𝐭𝐢𝐨𝐧 𝐮𝐬𝐢𝐧𝐠 <𝟏𝟎 𝐪𝐮𝐛𝐢𝐭𝐬 𝐰𝐢𝐭𝐡 𝐧𝐨𝐯𝐞𝐥 𝐚𝐫𝐜𝐡𝐢𝐭𝐞𝐜𝐭𝐮𝐫𝐞. What if size matters? A few years ago, people thought we’d need millions of qubits to break encryption. A few weeks ago, researchers said they’d only need 10,000… today Haiqu thinks that number is even smaller. With new state-of-the-art science calculations, Haiqu estimates that with 3-4 really big qubits (enormous), encryption will not stand a chance. USHER IN THE ERA OF QUANTUM ETERNITY WITH HAIQU BULQ’s NEW FAMILY-SIZED QUBITS. People keep making qubits small, but how do you even control something so tiny? You can’t even use your hands. INTRODUCING HAIQU BULQ: BIGGER, BETTER, AESTHETIC HARDWARE ARCHITECTURE UNVEILED BY HAIQU We envision a system where you can pull levers and ropes, there might be multiple people involved to execute a single operation. Comment MORE for a VR tour of one of our qubit’s interiors. “When people think of quantum, they think of microscopic things… tiny things.. So easy to lose. Haiqu believes reliability comes at scale. We will not have the most qubits, we’ll have the biggest. We just broke ground on three new data centres in Dubai, Bahrain, and Kuwait, each to store one qubit. The entire operation will be 100% oil powered, these are big, hot, qubits that need lots of energy. BULQ will be accessible via Haiqu SDK exclusively on ultra-wide monitors.” For investors, Haiqu has also added the entire crypto market ($2.44 trillion) to our Total Addressable Market since no encryption will stand a chance. #AprilFools