Recent benchmark evaluations reveal that the Vera Rubin NVL72 platform leads MLPerf Endpoints v1.0 inference and delivers twice the training performance of Blackwell Ultra. To support these processing capabilities, the company is deploying the OpenShell architecture and Sentry to manage and protect active agent environments. Developers are also utilizing Nemotron post-training workflows and specialized tools like the SOMA Retargeter to translate human motion for humanoid robots.
Key points
Hardware benchmark advances reflect an increasing industry focus on supporting agentic wo…: Hardware benchmark advances reflect an increasing industry focus on supporting agentic workflows that require continuous iterative reasoning.
Enterprise deployments are receiving dedicated governance frameworks to ensure secure exe…: Enterprise deployments are receiving dedicated governance frameworks to ensure secure execution across scaling AI agent environments.
Software ecosystem additions are broadening capabilities across diverse domains including…: Software ecosystem additions are broadening capabilities across diverse domains including scientific research, edge inference, and humanoid robotics.
Sources checked by our analysis pipeline: 1
NVIDIA Technical Blog
-
Data Center / CloudNVIDIA Vera Rubin NVL72 Leads Agentic Inference Performance on MLPerf Endpoints v1.0
-
Data Center / CloudNVIDIA Vera Rubin NVL72 Achieves up to 2x Blackwell Ultra Performance in MLPerf Training v6.1
-
Trustworthy AI / CybersecurityInside NVIDIA Sentry: Secure AI Agents at Runtime
-
Agentic AI / Generative AIHow NVIDIA OpenShell Reference Architecture Secures and Governs AI Agent Workloads at Scale
-
Simulation / Modeling / DesignDeveloping an AI Factory Digital Twin with an Agent-Assisted Workflow
Recent
Nov 16, 2026
Post-Training NVIDIA Nemotron to Increase Accuracy in Your Scientific Domain
Agentic AI is reshaping scientific research and innovation. Researchers communicate complex scientific hypotheses in natural language and let autonomous agents...
13 MIN READ
Nov 16, 2026
NVIDIA Vera Rubin NVL72 Leads Agentic Inference Performance on MLPerf Endpoints v1.0
Agentic workloads are demanding more from inference systems. AI agents carry growing conversation histories across turns and generate a mix of reasoning, tool...
8 MIN READ
Nov 16, 2026
NVIDIA Vera Rubin NVL72 Achieves up to 2x Blackwell Ultra Performance in MLPerf Training v6.1
Faced with nearly 30 disconnected source systems and tens of thousands of tables, enterprise teams often spend years on manual preparation before enabling analytics. To bypass this bottleneck, IFS Copperleaf engineered an internal platform that runs natural language queries directly over raw records rather than waiting for exhaustive data cleanup. This targeted modeling approach cuts months off deployment timelines while focusing high-precision modeling strictly on mission-critical business functions.
Enterprise organizations struggle to operationalize complex artificial intelligence systems because coordination challenges, rather than flawed models, derail most initiatives. Compounded handoffs between multiple models quickly degrade overall system reliability, creating significant architectural risks. To mitigate these failures, engineers must structure agent interactions around proven orchestration patterns that maintain operational control.
The enterprise platform handles 20 billion monthly transactions for leading institutional banks and asset managers that require rapid onboarding and dynamic data ingestion. Previous machine learning models demanded excessive customization and struggled with novel scenarios, prompting this shift to foundational AI frameworks. The resulting architecture significantly accelerates deployment cycles while enforcing enterprise-grade security controls across strictly segregated customer environments.