{"success":true,"data":{"pressRelease":{"id":"173025","rtpr_id":"nGNX6PQkqz-20261019","ticker":"NVDA","exchange":"NASDAQ","all_tickers":["NVDA"],"title":"NVIDIA Runs Record Simulation of Fault-Tolerant Quantum Computing at Jülich Supercomputing Centre","author":"Globe Newswire","published_at":"2026-10-19T18:00:44.100Z","article_body":"News Summary:\n* NVIDIA researchers are using JUPITER — the largest supercomputer in\nEurope, built at Jülich Supercomputing Centre — for a record-breaking\nfault-tolerant quantum computing simulation.\n* The 104-qubit simulation is the largest comprehensive simulation of a\nfault-tolerant quantum computing system to date, with no prior simulations\nmodeling quantum processors in such full detail.\n* The simulation validates a quantum computing approach that uses 90% fewer\nphysical qubits than previous methods, accelerating progress toward useful\napplications in drug discovery, financial modeling and materials science.\n* JUPITER simulations are providing the quantum industry with actionable\ninsight into the impact of noise on fault-tolerant quantum computing, enabling\nresearchers to refine designs before building hardware — shortening the path\nto viable quantum computers by years.\nBERLIN, Oct. 19, 2026 (GLOBE NEWSWIRE) -- NVIDIA today announced a\nrecord-breaking quantum computing simulation on JUPITER, Europe’s first\nexascale supercomputer, allowing researchers to fully explore and validate\nmagic state cultivation — a leading approach to fault-tolerant quantum\ncomputing that requires 90% fewer qubits than previous methods.\n\nRunning on a system of NVIDIA GH200 GPUs at Jülich Supercomputing Centre, the\n104-qubit simulation is the largest comprehensive simulation of a\nfault-tolerant quantum processor to date, modeling how a realistic quantum\nprocessor performs fault-tolerant computation at record scale.\n\nThat level of fidelity gives researchers the data and tools needed to predict\nthe performance of future designs before they are built — reducing reliance\non hardware trial and error, and accelerating the path to useful quantum\napplications in fields like drug discovery, financial modeling and materials\nscience.\n\n“Useful quantum computing will require fantastically complex, fault-tolerant\nsystems — and understanding how they behave at scale is essential to\nbuilding them,” said Jensen Huang, founder and CEO of NVIDIA. “Our results\nfrom Jülich show how quantum processors and GPU supercomputers can work\ntogether to tackle research that would otherwise be out of reach —\naccelerating the discovery of transformative quantum applications.”\n\n“Europe’s investment in JUPITER gives researchers the ability to\ninvestigate questions that were previously beyond reach,” said Kristel\nMichielsen, director of the Jülich Supercomputing Centre at the\nForschungszentrum Jülich facility. “Working with NVIDIA, we’re bringing\nthat capability to quantum computing — helping scientists make\nbetter-informed decisions about the designs most likely to deliver useful\napplications.”\n\nReducing the Resources Needed for Useful Quantum Computing\nBecause qubits operate at the limits of physics, the interactions needed to\ncontrol them introduce unavoidable, error-inducing noise. Computing in the\npresence of such noise — known as fault-tolerant quantum computing —\nrequires the use of magic states, a specific arrangement of qubits that is\ncontinually consumed during a fault-tolerant quantum computation.\n\nFault-tolerant quantum computers rely crucially on repeatedly producing magic\nstates — meaning that any optimization in how they are made drastically\nreduces the difficulty of building a useful quantum computer.\n\nBy simulating this process under realistic noise, NVIDIA researchers validated\na state-of-the art magic-state cultivation approach, giving researchers the\ntools needed to develop and test future improvements.\n\nEarlier, groundbreaking research from Jülich simulated 50 qubits under ideal,\nnoiseless conditions. This work simulates magic-state cultivation under\nrealistic hardware noise, helping researchers understand the impact of errors.\n\nFrom Research Development to Exascale Simulation\nSimulating magic-state cultivation helps researchers evaluate and improve\ndesigns before building quantum hardware. But the memory required can grow\nexponentially with the number of qubits, putting large-scale simulations\nbeyond the reach of conventional computing systems.\n\nAn NVIDIA GB200 NVL72 system at the NVIDIA Accelerated Quantum Research Center\n(https://www.nvidia.com/en-us/solutions/quantum-computing/accelerated-quantum-center/)\n(NVAQC) served as the development platform for the JUPITER project, enabling\nresearchers to scale tests from four GPUs to hundreds before moving to 8,192\nGPUs on JUPITER.\n\nResearchers used the center to accelerate the development and testing of their\nsimulation methods, compare results with JUPITER and explore how accuracy\nscaled with additional computing resources. This work helped reduce a\nweekslong simulation effort to days while preparing the team to use\nJUPITER’s exascale computing capacity effectively.\n\nThe project demonstrates how simulation will grow increasingly important and\ncomplex as quantum computing systems grow in size. It also shows the utility\nof accelerated computing clusters, like at NVAQC, in allowing researchers to\ndevelop and validate simulation methods before tapping into world-class\nsupercomputers for their most demanding calculations.\n\nThe simulations ran using NVIDIA cuQuantum\n(https://developer.nvidia.com/cuquantum-sdk) software to achieve highly\noptimized performance and scale, along with the NVIDIA CUDA-Q\n(https://developer.nvidia.com/cuda-q?size=n_6_n&sort-field=featured&sort-direction=desc)™\nplatform’s fault-tolerant logical layer to express the magic-state circuit.\n\nRead the GitHub blog\n(https://nvidia.github.io/cuda-quantum/blogs/blog/2026/10/19/jupiter-magic-state-simulation/)\non the JUPITER simulation.\n\nWatch the GTC Berlin keynote (https://www.nvidia.com/gtc/keynote/) from Huang\nand explore sessions (https://www.nvidia.com/en-eu/gtc/session-catalog/).\n\nAbout NVIDIA\nNVIDIA (https://www.nvidia.com/) (NASDAQ: NVDA) is the world leader in AI and\naccelerated computing.\n\nFor further information, contact:\nAaron Groff\nCorporate Communications\nNVIDIA Corporation\npress@nvidia.com\n\nCertain statements in this press release including, but not limited to,\nstatements as to: expectations with respect to growth, performance,\navailability, and benefits of NVIDIA’s products, services and technologies,\nand related trends and drivers; expectations with respect to NVIDIA’s third\nparty arrangements, including with its collaborators and partners;\nexpectations with respect to technology developments, and related trends and\ndrivers; projected market growth and trends; expectations with respect to AI\nand related industries; and other statements that are not historical facts are\nforward-looking statements within the meaning of Section 27A of the Securities\nAct of 1933, as amended, and Section 21E of the Securities Exchange Act of\n1934, as amended, which are subject to the “safe harbor” created by those\nsections based on management’s beliefs and assumptions and on information\ncurrently available to management and are subject to risks and uncertainties\nthat could cause results to be materially different than expectations.\nImportant factors that could cause actual results to differ materially\ninclude: global economic and political conditions; NVIDIA’s reliance on\nthird parties to manufacture, assemble, package and test NVIDIA’s products;\nthe impact of technological development and competition; development of new\nproducts and technologies or enhancements to NVIDIA’s existing products and\ntechnologies; market acceptance of NVIDIA’s products or NVIDIA’s\npartners’ products; design, manufacturing or software defects; changes in\nconsumer preferences or demands; changes in industry standards and interfaces;\nunexpected loss of performance of NVIDIA’s products or technologies when\nintegrated into systems; NVIDIA’s ability to realize the potential benefits\nof business investments or acquisitions; and changes in applicable laws and\nregulations, as well as other factors detailed from time to time in the most\nrecent reports NVIDIA files with the Securities and Exchange Commission, or\nSEC, including, but not limited to, its Annual Report on Form 10-K and\nQuarterly Reports on Form 10-Q. Copies of reports filed with the SEC are\nposted on NVIDIA’s website and are available from NVIDIA without charge.\nThese forward-looking statements are not guarantees of future performance and\nspeak only as of the date hereof, and, except as required by law, NVIDIA\ndisclaims any obligation to update these forward-looking statements to reflect\nfuture events or circumstances.\n\n© 2026 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, and\nother NVIDIA product and service names are trademarks and/or registered\ntrademarks of NVIDIA Corporation in the U.S. and other countries and regions.\nOther company, product, and service names may be trademarks of the respective\ncompanies with which they are associated.\n\nA photo accompanying this announcement is available at\nhttps://www.globenewswire.com/NewsRoom/AttachmentNg/f89cad77-76b7-4c56-a133-3ec9e0441d34\n\n\n\n(https://www.globenewswire.com/NewsRoom/AttachmentNg/ad3b9553-6933-40b1-adfd-2ed199cd141c)\nJUPITER Supercomputer \n(https://www.globenewswire.com/NewsRoom/AttachmentNg/f89cad77-76b7-4c56-a133-3ec9e0441d34/en)\nNVIDIA researchers are using JUPITER — the largest supercomputer in Europe,\nbuilt at Jülich Supercomputing Centre — for a record-breaking\nfault-tolerant quantum computing simulation.\n\n\nGlobeNewswire, Inc. 2026","article_body_html":"","raw_payload":{"data":{"id":"nGNX6PQkqz-20261019","title":"NVIDIA Runs Record Simulation of Fault-Tolerant Quantum Computing at Jülich Supercomputing Centre","author":"Globe Newswire","ticker":"NVDA","created":"2026-10-19T18:00:44.100Z","tickers":["NVDA"],"exchange":"NASDAQ","article_body":"News Summary:\n* NVIDIA researchers are using JUPITER — the largest supercomputer in\nEurope, built at Jülich Supercomputing Centre — for a record-breaking\nfault-tolerant quantum computing simulation.\n* The 104-qubit simulation is the largest comprehensive simulation of a\nfault-tolerant quantum computing system to date, with no prior simulations\nmodeling quantum processors in such full detail.\n* The simulation validates a quantum computing approach that uses 90% fewer\nphysical qubits than previous methods, accelerating progress toward useful\napplications in drug discovery, financial modeling and materials science.\n* JUPITER simulations are providing the quantum industry with actionable\ninsight into the impact of noise on fault-tolerant quantum computing, enabling\nresearchers to refine designs before building hardware — shortening the path\nto viable quantum computers by years.\nBERLIN, Oct. 19, 2026 (GLOBE NEWSWIRE) -- NVIDIA today announced a\nrecord-breaking quantum computing simulation on JUPITER, Europe’s first\nexascale supercomputer, allowing researchers to fully explore and validate\nmagic state cultivation — a leading approach to fault-tolerant quantum\ncomputing that requires 90% fewer qubits than previous methods.\n\nRunning on a system of NVIDIA GH200 GPUs at Jülich Supercomputing Centre, the\n104-qubit simulation is the largest comprehensive simulation of a\nfault-tolerant quantum processor to date, modeling how a realistic quantum\nprocessor performs fault-tolerant computation at record scale.\n\nThat level of fidelity gives researchers the data and tools needed to predict\nthe performance of future designs before they are built — reducing reliance\non hardware trial and error, and accelerating the path to useful quantum\napplications in fields like drug discovery, financial modeling and materials\nscience.\n\n“Useful quantum computing will require fantastically complex, fault-tolerant\nsystems — and understanding how they behave at scale is essential to\nbuilding them,” said Jensen Huang, founder and CEO of NVIDIA. “Our results\nfrom Jülich show how quantum processors and GPU supercomputers can work\ntogether to tackle research that would otherwise be out of reach —\naccelerating the discovery of transformative quantum applications.”\n\n“Europe’s investment in JUPITER gives researchers the ability to\ninvestigate questions that were previously beyond reach,” said Kristel\nMichielsen, director of the Jülich Supercomputing Centre at the\nForschungszentrum Jülich facility. “Working with NVIDIA, we’re bringing\nthat capability to quantum computing — helping scientists make\nbetter-informed decisions about the designs most likely to deliver useful\napplications.”\n\nReducing the Resources Needed for Useful Quantum Computing\nBecause qubits operate at the limits of physics, the interactions needed to\ncontrol them introduce unavoidable, error-inducing noise. Computing in the\npresence of such noise — known as fault-tolerant quantum computing —\nrequires the use of magic states, a specific arrangement of qubits that is\ncontinually consumed during a fault-tolerant quantum computation.\n\nFault-tolerant quantum computers rely crucially on repeatedly producing magic\nstates — meaning that any optimization in how they are made drastically\nreduces the difficulty of building a useful quantum computer.\n\nBy simulating this process under realistic noise, NVIDIA researchers validated\na state-of-the art magic-state cultivation approach, giving researchers the\ntools needed to develop and test future improvements.\n\nEarlier, groundbreaking research from Jülich simulated 50 qubits under ideal,\nnoiseless conditions. This work simulates magic-state cultivation under\nrealistic hardware noise, helping researchers understand the impact of errors.\n\nFrom Research Development to Exascale Simulation\nSimulating magic-state cultivation helps researchers evaluate and improve\ndesigns before building quantum hardware. But the memory required can grow\nexponentially with the number of qubits, putting large-scale simulations\nbeyond the reach of conventional computing systems.\n\nAn NVIDIA GB200 NVL72 system at the NVIDIA Accelerated Quantum Research Center\n(https://www.nvidia.com/en-us/solutions/quantum-computing/accelerated-quantum-center/)\n(NVAQC) served as the development platform for the JUPITER project, enabling\nresearchers to scale tests from four GPUs to hundreds before moving to 8,192\nGPUs on JUPITER.\n\nResearchers used the center to accelerate the development and testing of their\nsimulation methods, compare results with JUPITER and explore how accuracy\nscaled with additional computing resources. This work helped reduce a\nweekslong simulation effort to days while preparing the team to use\nJUPITER’s exascale computing capacity effectively.\n\nThe project demonstrates how simulation will grow increasingly important and\ncomplex as quantum computing systems grow in size. It also shows the utility\nof accelerated computing clusters, like at NVAQC, in allowing researchers to\ndevelop and validate simulation methods before tapping into world-class\nsupercomputers for their most demanding calculations.\n\nThe simulations ran using NVIDIA cuQuantum\n(https://developer.nvidia.com/cuquantum-sdk) software to achieve highly\noptimized performance and scale, along with the NVIDIA CUDA-Q\n(https://developer.nvidia.com/cuda-q?size=n_6_n&sort-field=featured&sort-direction=desc)™\nplatform’s fault-tolerant logical layer to express the magic-state circuit.\n\nRead the GitHub blog\n(https://nvidia.github.io/cuda-quantum/blogs/blog/2026/10/19/jupiter-magic-state-simulation/)\non the JUPITER simulation.\n\nWatch the GTC Berlin keynote (https://www.nvidia.com/gtc/keynote/) from Huang\nand explore sessions (https://www.nvidia.com/en-eu/gtc/session-catalog/).\n\nAbout NVIDIA\nNVIDIA (https://www.nvidia.com/) (NASDAQ: NVDA) is the world leader in AI and\naccelerated computing.\n\nFor further information, contact:\nAaron Groff\nCorporate Communications\nNVIDIA Corporation\npress@nvidia.com\n\nCertain statements in this press release including, but not limited to,\nstatements as to: expectations with respect to growth, performance,\navailability, and benefits of NVIDIA’s products, services and technologies,\nand related trends and drivers; expectations with respect to NVIDIA’s third\nparty arrangements, including with its collaborators and partners;\nexpectations with respect to technology developments, and related trends and\ndrivers; projected market growth and trends; expectations with respect to AI\nand related industries; and other statements that are not historical facts are\nforward-looking statements within the meaning of Section 27A of the Securities\nAct of 1933, as amended, and Section 21E of the Securities Exchange Act of\n1934, as amended, which are subject to the “safe harbor” created by those\nsections based on management’s beliefs and assumptions and on information\ncurrently available to management and are subject to risks and uncertainties\nthat could cause results to be materially different than expectations.\nImportant factors that could cause actual results to differ materially\ninclude: global economic and political conditions; NVIDIA’s reliance on\nthird parties to manufacture, assemble, package and test NVIDIA’s products;\nthe impact of technological development and competition; development of new\nproducts and technologies or enhancements to NVIDIA’s existing products and\ntechnologies; market acceptance of NVIDIA’s products or NVIDIA’s\npartners’ products; design, manufacturing or software defects; changes in\nconsumer preferences or demands; changes in industry standards and interfaces;\nunexpected loss of performance of NVIDIA’s products or technologies when\nintegrated into systems; NVIDIA’s ability to realize the potential benefits\nof business investments or acquisitions; and changes in applicable laws and\nregulations, as well as other factors detailed from time to time in the most\nrecent reports NVIDIA files with the Securities and Exchange Commission, or\nSEC, including, but not limited to, its Annual Report on Form 10-K and\nQuarterly Reports on Form 10-Q. Copies of reports filed with the SEC are\nposted on NVIDIA’s website and are available from NVIDIA without charge.\nThese forward-looking statements are not guarantees of future performance and\nspeak only as of the date hereof, and, except as required by law, NVIDIA\ndisclaims any obligation to update these forward-looking statements to reflect\nfuture events or circumstances.\n\n© 2026 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, and\nother NVIDIA product and service names are trademarks and/or registered\ntrademarks of NVIDIA Corporation in the U.S. and other countries and regions.\nOther company, product, and service names may be trademarks of the respective\ncompanies with which they are associated.\n\nA photo accompanying this announcement is available at\nhttps://www.globenewswire.com/NewsRoom/AttachmentNg/f89cad77-76b7-4c56-a133-3ec9e0441d34\n\n\n\n(https://www.globenewswire.com/NewsRoom/AttachmentNg/ad3b9553-6933-40b1-adfd-2ed199cd141c)\nJUPITER Supercomputer \n(https://www.globenewswire.com/NewsRoom/AttachmentNg/f89cad77-76b7-4c56-a133-3ec9e0441d34/en)\nNVIDIA researchers are using JUPITER — the largest supercomputer in Europe,\nbuilt at Jülich Supercomputing Centre — for a record-breaking\nfault-tolerant quantum computing simulation.\n\n\nGlobeNewswire, Inc. 2026"},"type":"article","timestamp":"2026-10-19T18:00:44.137473151Z","server_sent_at_ms":1792432844137},"received_at":"2026-10-19T18:00:44.191Z","source_url":null},"analysis":{"id":"161134","press_release_id":"173025","analysis_json":{"industry":{"label":"Semiconductors & Semiconductor Equipment","sector":"Information Technology"},"redFlags":[],"eventType":"product_launch","narrative":"NVIDIA announced a record 104-qubit simulation of fault-tolerant quantum computing run on JUPITER, Europe's first exascale supercomputer at the Jülich Supercomputing Centre.\n\nThe work validated a magic-state cultivation approach that uses 90% fewer physical qubits than prior methods, and demonstrated 8,192 NVIDIA GPUs working in tandem via GH200 hardware and the CUDA-Q/cuQuantum software stack.\n\nWhile not a financial event, the milestone strengthens NVIDIA's positioning as the accelerated-computing backbone for quantum research and shortens the industry's path to useful quantum applications.","sentiment":"bullish","agentHooks":{"shouldPost":false,"suggestedAngle":"NVIDIA demonstrates GPU supercomputing as the simulation backbone for fault-tolerant quantum computing at record 104-qubit scale."},"keyFigures":{"customDimensions":{"gpus_on_jupiter":8192,"qubits_simulated":104,"physical_qubit_reduction":"90%"}},"quotedText":"Our results from Jülich show how quantum processors and GPU supercomputers can work together to tackle research that would otherwise be out of reach","namedEntities":{"people":[{"name":"Jensen Huang","role":"founder and CEO of NVIDIA"},{"name":"Kristel Michielsen","role":"director of the Jülich Supercomputing Centre, Forschungszentrum Jülich"},{"name":"Aaron Groff","role":"Corporate Communications, NVIDIA"}],"products":["JUPITER supercomputer","NVIDIA GH200","NVIDIA GB200 NVL72","NVIDIA CUDA-Q","NVIDIA cuQuantum","NVAQC (NVIDIA Accelerated Quantum Research Center)"],"companies":[{"name":"NVIDIA Corporation","ticker":"NVDA","relationship":"filer"},{"name":"Jülich Supercomputing Centre / Forschungszentrum Jülich","relationship":"research partner"}],"dollarAmounts":[]},"materialImpact":{"score":2,"reasoning":"A research milestone showcasing NVIDIA's GH200 GPUs, GB200 NVL72, CUDA-Q and cuQuantum in record-scale quantum simulation. It reinforces NVIDIA's platform positioning in quantum-era accelerated computing but carries no direct near-term revenue or financial impact."},"tickerRelevance":{"others":[],"primary":"NVDA"},"globalImportance":35,"audienceRelevance":65,"eventTypeSecondary":["operations_update"],"importanceComponents":{"tickerTier":"mega-cap","eventGravity":"research-milestone","retailFavoriteBoost":true,"financialMateriality":"low","technologySignificance":"high"}},"event_type":"product_launch","event_type_secondary":["operations_update"],"sentiment":"bullish","material_impact_score":2,"narrative":"NVIDIA announced a record 104-qubit simulation of fault-tolerant quantum computing run on JUPITER, Europe's first exascale supercomputer at the Jülich Supercomputing Centre.\n\nThe work validated a magic-state cultivation approach that uses 90% fewer physical qubits than prior methods, and demonstrated 8,192 NVIDIA GPUs working in tandem via GH200 hardware and the CUDA-Q/cuQuantum software stack.\n\nWhile not a financial event, the milestone strengthens NVIDIA's positioning as the accelerated-computing backbone for quantum research and shortens the industry's path to useful quantum applications.","key_figures":{"customDimensions":{"gpus_on_jupiter":8192,"qubits_simulated":104,"physical_qubit_reduction":"90%"}},"named_entities":{"people":[{"name":"Jensen Huang","role":"founder and CEO of NVIDIA"},{"name":"Kristel Michielsen","role":"director of the Jülich Supercomputing Centre, Forschungszentrum Jülich"},{"name":"Aaron Groff","role":"Corporate Communications, NVIDIA"}],"products":["JUPITER supercomputer","NVIDIA GH200","NVIDIA GB200 NVL72","NVIDIA CUDA-Q","NVIDIA cuQuantum","NVAQC (NVIDIA Accelerated Quantum Research Center)"],"companies":[{"name":"NVIDIA Corporation","ticker":"NVDA","relationship":"filer"},{"name":"Jülich Supercomputing Centre / Forschungszentrum Jülich","relationship":"research partner"}],"dollarAmounts":[]},"model_name":"glm-5.3-flashx","prompt_hash":"sha256:5c812bd9e28b810d","schema_hash":"sha256:05005c02d9cffac9","created_at":"2026-10-19T18:00:49.362Z","global_importance":35,"audience_relevance":65,"importance_components":{"tickerTier":"mega-cap","eventGravity":"research-milestone","retailFavoriteBoost":true,"financialMateriality":"low","technologySignificance":"high"}},"durationMs":5159,"modelName":"glm-5.3-flashx"}}