TSTickerStarGet alerts

NVIDIA Announces Nemotron for Science, Rubin HPC GPU and Expanded Vera CPU Platform at SC26

NVIDIA's Ian Buck, vice president of hyperscale and high-performance computing, kicked off the company's week at SC26 in Chicago with a special address announcing Nemotron for Science, the Rubin HPC GPU and an expanded Vera CPU platform for high-performance computing. The article contains no reported financial results, no revenue or EPS figures, and no financial guidance. NVIDIA and IBM, with Fermilab, also announced plans for a quantum computing co-design platform, and a simulation on Jülich's JUPITER supercomp...

Share this story

Key takeaways

  • Buck announced Nemotron for Science, combining the Nemotron 3.5 Super VL open model with a new open chemistry dataset and post-training recipes for life sciences and chemistry; researchers can run it locally on a DGX Spark, a workstation or their own supercomputer.

  • At Argonne National Laboratory, an agent built with the toolkit took a single prompt, planned and tested hypotheses, chose which simulations to run and identified two new binding sites.

  • Rubin HPC is a new GPU in the Vera Rubin platform supporting precisions from FP64 to FP4 with sparse-data acceleration and seventh-generation NVIDIA NVLink; it has more than 50,000 CUDA cores, about three times as many as Hopper, and is expected in 2028.

  • DKRZ, the German Climate Computing Center, selected Vera CPUs for its HLRE-5 supercomputer, expected in early 2028, joining Vera systems already announced at Texas Advanced Computing Center.

  • NVIDIA, IBM and Fermilab announced plans for a quantum computing co-design platform, pairing NVIDIA DGX B200 systems with IBM quantum computers including IBM Quantum Nighthawk.

  • A new CUDA-Q decoding algorithm, TCBP-JOBS, cuts error-correction research computation from 12 weeks to 45 minutes, with the goal of cutting decoding latency from about 300 microseconds to 10.

  • NVIDIA showed one of the first deployments of Photonics co-packaged optics switches at Lambda, said to be 15x more reliable on bit errors and link flaps.

Key metrics

CUDA-X libraries offered
more than 1,000
Rubin HPC CUDA cores
more than 50,000 CUDA cores (about three times as many as Hopper)
Rubin HPC expected availability
2028
Vera CPU selection for DKRZ HLRE-5 supercomputer
selected; expected in early 2028
Magic state cultivation reduction in operations for one logical step
tenfold reduction
CUDA-Q TCBP-JOBS error-correction research computation time
from 12 weeks to 45 minutes
Quantum decoding latency goal
from about 300 microseconds to 10
Nsight AI Porting Agent WRF GPU speedup
more than 5x using about 150 million tokens

Primary source

TickerStar generated this brief from an official company, investor-relations, earnings-call, or regulatory source. Review the primary material before making an investment decision.

Open Investor relations

Follow NVDA in real time

Get material earnings, filing, IR, and earnings-call alerts by email.

Create a free alert