NVIDIA and OpenAI are building the largest data center ever at gigawatt scale.
“When it's fully built out, it will be the largest data center ever built.”
NVIDIA
125 tracked signals on NVIDIA.
Competing open letters expose industry fracture over open-weight AI models and development pacing
“We request that the U.S. government support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development.”
NVIDIA is now running a wide range of frontier AI models.
“as a company I rather for us to help everybody succeed instead of taking a slice”
NVIDIA launches NVLink Fusion to enhance custom AI infrastructure.
NVIDIA is shaping infrastructure for agentic AI with a new platform called Vera Rubin.
“The value gets generated in the computing and the transformation of the actual data into a calculation.”
TensorRT Edge-LLM achieves 6.4x faster MLPerf benchmark on Jetson AGX Thor.
Integration of NVIDIA Resiliency Extension enhances fault tolerance in distributed training on Amazon EKS.
NVIDIA introduces cuTile Rust for GPU kernel authoring in Rust.
DocuSign partners with NVIDIA to tackle unstructured agreement data.
“there's $2 trillion captured in this agreement negotiated value that no one capitalizes”
NVIDIA introduces NVLink Fusion for AI infrastructure.
NVIDIA's Vera Rubin platform prioritizes power efficiency for AI workloads.
“Power is a defining constraint for AI factories.”
NVIDIA NVLink 6 enhances multi-layer resiliency for AI factories.
NVIDIA DSX aims to create more energy-efficient AI factories.
MoE architecture is a major trend in efficient AI model training.
NVIDIA announced Bionemo inference runtime to enhance biomolecular structure prediction.
“Get started today with the open-source Bonimmo inference runtime.”
NVIDIA introduces BioNeMo Inference Runtime for efficient biomolecular structure prediction.
NVIDIA introduces EPD disaggregation for multimodal model optimization.
Adobe Premiere is optimized for NVIDIA TRT RTX acceleration.
“There's also a new feature within Premiere called object mask tool, which allows us to track an object with one click, significantly faster than our previous version of Premiere.”
NVIDIA is introducing CUDA Rust for native GPU programming.
BSC AI Factory exemplifies Europe's AI infrastructure initiative.
“AI Factory is a project by Europe for Europe.”
NVIDIA NemoClaw enables memory-driven AI agents for enterprise work.
NVIDIA Cosmos 3 enables a streamlined Physical AI model factory on Amazon SageMaker HyperPod.
NVIDIA introduces the PAIR Virtual Inference Router to enhance AI agent collaboration.
Speculative decoding can accelerate LLM inference while maintaining accuracy.
Alibaba releases open weights for 2.4T-parameter Qwen3.8-Max, its largest open-weight model
Meta releases Muse Glimmer, a 30B open-weight model delivering 20K tokens/sec on a single GPU for local agentic AI.
“Meta returns to the open source ecosystem with the release of Muse Glimmer, a 30B open-weight dense model with a 120K+ context window built for local AI agentic work.”
NVIDIA Rubin GPU architecture is purpose-built for agentic AI workloads at scale
“What began as discrete AI model training and human-facing chat interfaces has evolved into always-on AI factories dedicated to producing intelligence at scale.”
NVIDIA Cosmos is an open foundation model generating synthetic data for physical AI training.
“For physical AI, compute is data.”
NVIDIA unveils Vera Rubin in full production, RTX Spark AI PCs, and an enterprise agent toolkit at GTC Taipei.
“Vera Rubin is the most ambitious endeavor in the history of our company.”
NVIDIA DSX is a reference design for building gigawatt-scale AI factories at maximum efficiency and profitability.
“Today's AI factories over provision power by up to 40%.”
NVIDIA launched Cosmos 3, Nemotron 3 Ultra, and previewed the RTX Spark superchip at Computex.
NVIDIA releases Cosmos 3, billed as the first open omni-model for physical AI reasoning and action.
NVIDIA marks 10 years of DGX, evolving from a single system into the blueprint for modern AI factories.
“And now with the NVIDIA Vera Rubin rack-scale system, we are seeing up to a 10x reduction in inference token cost, crucial for today's agentic AI workloads.”
Salesforce will become an AI company, building private custom models on NVIDIA's platform.
“Salesforce is going to be an AI company.”
NVIDIA emphasizes the importance of safety in innovation.
“Safety is paramount.”
NVIDIA FLARE improves scaling of federated learning across multiple platforms.
Full-stack NIM optimizations can deliver 2.5x more users on Nemotron 3 Ultra.
CUDA Toolkit 13.4 introduces support for Windows on Arm.
Deploying multi-step reasoning models on NVIDIA Jetson becomes feasible.
NVIDIA and CrowdStrike partner on SafeMind to build domain-specialized cyber defense AI
“you must have the ability to fine-tune to post-train in the context of safe mind um create an AI that is super good at a particular domain”
NVIDIA BioNeMo NIM microservices now integrate with Claude for agentic protein structure prediction
“Agentic AI is changing how research is done. AI scientists can read papers, propose hypotheses, call models, and determine which experiments to prioritize next.”
NVIDIA NVLink Fusion enables NVHBM memory for custom XPU accelerators at hyperscale
NVIDIA Vera Rubin NVL72 production racks are now shipping with fully automated assembly.
NVIDIA Vera Rubin and Blackwell set new performance-per-watt benchmarks for agentic AI workloads
“across 100 trillion tokens of real-world usage, OpenRouter's State of AI report found that average prompt tokens per request grew roughly fourfold”
NVIDIA Groq 3 LPX accelerator enables ultrafast interactive inference on Vera Rubin NVL72 at long context
“NVIDIA Vera Rubin NVL72, the most versatile machine ever built, delivering high throughput and interactivity across the widest range of AI workloads—from small to large models, both open and closed.”
NVIDIA Nemotron 3.5 Lightning brings 4x agentic throughput to SageMaker JumpStart on a single GPU
NVIDIA launches Nemotron 3.5 Lightning, a 30B MoE model optimized for agentic execution layers
“Long-running AI agents spend most of their time on high-volume execution: tool calls, result validation, and subagent delegation.”
New Open Secure AI Alliance launched with Palantir to keep enterprise AI outputs under company ownership.
“when you bring employees into your company and they do work on behalf of your company, you own that IP. It should be the same thing with the intelligence that you bring in”
NVIDIA RTX Spark is a 1-petaflop superchip for running advanced AI agents locally on laptops
“The reinvention of the personal computer has begun and it starts with the RTX Spark”
NVIDIA's Rubin GPU architecture kills megakernels, ending a major inference optimization research direction
“the GPU is designed in such a way that it kills mega kernels. So it seems like that entire research field won't be continued.”
Multi-model routing in AI agents cuts latency 50% and tokens 25% with no quality loss
“Intelligence isn't one size-fits-all.”
Autonomous AI agents now span the full EDA stack from RTL design to chip signoff
“Cadence's autonomous AI engineer compresses RTL development from weeks to hours, and drives the flow itself.”
Agentic AI demands a new enterprise data category—agent context—forcing storage to become systems of context.
“Storage needs to evolve. It needs to move from being what we've always known as a system of record, to now a system of context.”
NVIDIA frames AI-generated virtual worlds as the training ground for physical AI and robotics.
“AI will generate amazing virtual worlds. And in these virtual worlds, the next era of AI, robotics will begin.”
NVIDIA introduces World-Action Models (WAMs), robot policies fine-tuned from pretrained world/video models to act.
“WAM World-Action Model: a policy that starts from a pretrained world-model or video”
Mistral and NVIDIA's Nemotron Coalition will train and release a new open-source frontier model.
“The benefit for everyone involved, really, is that we will have a new open source frontier model that everyone can build off on.”
NVIDIA DOCA GPUNetIO enables GPUs to control networking directly, eliminating CPU bottlenecks
“GPU applications increasingly need networking and data movement to behave like first-class GPU-controlled operations rather than host-driven services.”
NVIDIA introduces Green Contexts for fine-grained GPU resource sharing between concurrent workloads
“Controlling how GPU resources are shared between them remains difficult.”
NVIDIA releases domain-specific DOCA Agent Skills for BlueField infrastructure development
“AI agents are becoming a standard part of development workflows, but general-purpose agents weren't built with specialized infrastructure software such as NVIDIA DOCA in mind.”
NVIDIA launches cuObject and SCADA Server SDK for faster AI storage access