Lifestyle
NVIDIA Unveils Rubin HPC GPU: Over 50,000 CUDA Cores per Chip, Expected in 2028
At SC26, NVIDIA announced Rubin HPC, a new member of the Vera Rubin platform aimed at scientific, engineering and analytics simulation. Each GPU has over 50,000 CUDA cores, with availability expected in 2028.
News date: About 4 min read
Contents
On November 16, 2026, NVIDIA announced Rubin HPC at SC26. HPC stands for high-performance computing: using large amounts of computing resources for big scientific and engineering calculations. Rubin HPC is a new GPU in the Vera Rubin platform, built for simulation and data processing in science, engineering and analytics. NVIDIA says the product is expected to ship in 2028 and is still at the preview stage. For research and engineering teams, simulation speed affects how quickly they can iterate from hypothesis to verified result.
What Is Rubin HPC?
According to the NVIDIA blog, Rubin HPC is a GPU designed for simulation and data processing. NVIDIA notes that AI agents (AI programs that carry out tasks automatically on a person's behalf) are helping scientists and engineers explore more candidate materials, molecules and designs, which calls for more simulation to validate those ideas. NVIDIA describes the GPU as a "balanced design" that scales data movement and compute together rather than only raising compute. CUDA cores are the basic units in NVIDIA GPUs that perform parallel computation. NVLink is the interconnect technology that lets multiple GPUs exchange data with one another.
| Aspect | NVIDIA's description |
|---|---|
| Product positioning | A new GPU in the Vera Rubin platform, for simulation and data processing |
| CUDA core count | More than 50,000 per GPU (NVIDIA calls this a record) |
| Data supply | Sparse-data acceleration that keeps cores supplied with data to compute |
| GPU interconnect | Seventh-generation NVIDIA NVLink |
| Software | CUDA-X libraries, offering optimized algorithms for simulation and analytics |
| Availability | Expected in 2028 |
Which Workloads Benefit?
NVIDIA says workloads such as climate modeling, engineering simulation, genomics analysis and financial risk share a similar pattern: the program pulls a few scattered values from a huge dataset, computes a result, and repeats this at large scale. This situation, where only scattered values are needed from a large amount of data, is what sparse data means. NVIDIA says Rubin HPC can accelerate these recurring compute patterns, including sparse matrix operations; stencil operations, which compute each point from the values of neighboring grid points; and sorting and reduction, which aggregate many values into a single result.
What Does It Mean for Research and Engineering Workflows?
NVIDIA says that when simulations invoked by AI agents run faster, the loop of forming a hypothesis, running a simulation and revising can be shortened. Researchers and engineers can then decide their next step sooner and focus lab work on the most promising candidates. Ian Buck, NVIDIA's vice president of hyperscale and high-performance computing, said: "For agents tackling science and engineering, faster simulation means faster discovery." All of this is the vendor's own account; the announcement provides no measured performance data.
Frequently Asked Questions
Can I buy Rubin HPC now?
No. NVIDIA says Rubin HPC is expected to ship in 2028; for now it is only a preview announcement.
How are Rubin HPC and Vera Rubin related?
According to NVIDIA, Rubin HPC is a new GPU in the Vera Rubin platform, dedicated to simulation and data processing in science, engineering and analytics.
Does the announcement give pricing or regional availability?
No. NVIDIA's announcement does not mention pricing or regional availability details.
What role does CUDA-X play?
NVIDIA says the CUDA-X libraries provide optimized algorithm implementations for simulation and analytics, for use by AI agents, and can be paired with Rubin HPC.
Browse the latest news in this topic
Found this guide helpful? Share it with someone who needs it
Lifestyle
AMD Says Lux Supercomputer Is Open to Scientific Users at Oak Ridge, Ranked No. 14 on TOP500 and Built on MI355X GPUs
AMD announced that the AMD-powered, HPE-built Lux supercomputer is now open to scientific users at the US Department of Energy's Oak Ridge National Laboratory and is the first dedicated AI Factory for the Genesis Mission. AMD encourages interested users to contact OLCF.
Lifestyle
Google report: attackers use AI to find vulnerabilities from software updates and build exploits; Google introduces AI Threat Defense
Google Threat Intelligence Group (GTIG) said on November 16, 2026 that its AI Threat Tracker report found attackers using autonomous AI frameworks to analyze software updates, identify patched flaws and build exploits. Google also introduced AI Threat Defense, combining Gemini, Wiz, CodeMender and Mandiant, but gave no availability stage or pricing.
Lifestyle
Google Cloud Publishes MLPerf Training v6.1 Results: 8,192 Ironwood TPUs Hit Target in 1.70 Minutes, Plus the First Agentic Reinforcement Learning Test
Google Cloud says it submitted 10 results to the MLPerf Training v6.1 AI benchmark using Ironwood TPUs and NVIDIA Grace Blackwell GPUs, covering pretraining and the first agentic reinforcement learning benchmark. This is a results announcement, not a product launch; all figures are Google's own.
Lifestyle
NVIDIA Vera Rubin Debuts in MLPerf: Claims a Sweep of Training v6.1, With Over 10x Throughput per Watt on Kimi K3 (Provisional)
NVIDIA says its Vera Rubin platform posted leading results in MLPerf Training v6.1 and the debut MLPerf Endpoints, though the Vera Rubin Endpoints results are still pending MLCommons verification.
Sources
- NVIDIA Announces Rubin HPC — A New GPU to Accelerate Agents for Science, Engineering and AnalyticsChecked:
Compiled from the references and verification dates listed above; fares, schedules, product features, and terms can change, so please confirm the latest details on the relevant official website before travelling, booking, or purchasing.