Quantum
NVIDIA Runs Simulation of Fault-Tolerant Quantum Computing at Jülich Supercomputing Centre
Publication: The Quantum Insider
Author: Matt Swayne
Source published: 19/10/2026, 19:02:39 UTC
Briefing created: 19/10/2026, 20:06:01 UTC
Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 19/10/2026, 20:04:25 UTC.
Briefings
NVIDIA successfully simulated a 104-qubit fault-tolerant quantum processor on Europe's JUPITER supercomputer to evaluate magic-state cultivation under realistic noise. The simulation utilized 8,192 GPUs and NVIDIA's quantum software to validate a method that could reduce qubit requirements by 90% compared to prior approaches. Jensen Huang and Kristel Michielsen highlighted how this exascale simulation provides essential data for predicting future processor designs before physical hardware is built.
Key points
- NVIDIA simulated a 104-qubit fault-tolerant quantum processor on JUPITER using 8,192 GPUs and NVIDIA's quantum software.
- The simulation evaluated magic-state cultivation under realistic noise, potentially reducing qubit requirements by 90% compared to previous methods.
Context
This exascale simulation on JUPITER allows researchers to validate magic-state cultivation methods under realistic noise, reducing reliance on hardware trial and error. The project demonstrates how simulation scales with additional computing resources, accelerating the path to useful quantum applications in drug discovery and materials science.
Limitations
Evidence is limited to the NVIDIA simulation on JUPITER; earlier 50-qubit ideal noiseless simulations are not included. Results are based on NVIDIA's internal assessment and may vary with future hardware improvements or noise characteristics.
Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 19/10/2026, 20:06:01 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.
View archived briefing on IPFS
bafybeifv5r7gtthfsdbwbna3t4fghwdl2jfcqvcgeodn2sth3xpl4ce6py