NVIDIA Dev Blog

Teams building AI applications need a repeatable path from a model configuration to a working inference endpoint. NVIDIA Dynamo is a distributed inference... Teams building AI applications need a repeatable path from a model configuration to a working inference endpoint. NVIDIA Dynamo is a distributed inference framework that coordinates model serving across GPU workers. Deploying it on Kubernetes is only a first step. A successful model deployment requires more: GPU drivers, resource allocation, scheduling, networking, and monitoring must all work… Source

AI is expanding beyond the data center as developers build a new generation of agents to act at the edge: monitoring driver attention and cabin conditions,... AI is expanding beyond the data center as developers build a new generation of agents to act at the edge: monitoring driver attention and cabin conditions, helping robots understand their surroundings and plan tasks, and coordinating fleets of machines on production lines. These physical AI applications rely on models that perceive, reason, and act within tight memory, power, and latency budgets. Source

Building an autonomous agent requires a model, a harness, and domain-specific tools. To deploy an agent more securely, developers need a runtime that enforces... Building an autonomous agent requires a model, a harness, and domain-specific tools. To deploy an agent more securely, developers need a runtime that enforces policies, along with optimization and observability tools for reliable execution. For production agents, the challenge lies in making these agents capable and safe to operate with access to real enterprise systems and data. Source

NVIDIA GPUDirect Storage (GDS) accelerates GPU storage access in on-premises and cloud environments. It provides a robust, portable, and efficient data path... NVIDIA GPUDirect Storage (GDS) accelerates GPU storage access in on-premises and cloud environments. It provides a robust, portable, and efficient data path between NVIDIA GPUs and storage targets to increase bandwidth, minimize latency, and reduce CPU and memory utilization across a broad range of platforms, application frameworks, and workload patterns. Configuring NVIDIA GPUDirect Storage… Source

Atomistic simulations help researchers understand how molecules and materials behave, from chemical reactions and phase transitions to mechanical failure. These... Atomistic simulations help researchers understand how molecules and materials behave, from chemical reactions and phase transitions to mechanical failure. These simulations come with a cost-accuracy trade-off. Quantum-mechanical methods such as density functional theory (DFT) can provide accurate energies and forces, but their computational cost often limits their use for large systems or long… Source

Many AI workloads need only a fraction of a GPU. Giving each one an entire device can leave capacity idle; sharing GPUs lets platform teams serve more workloads... Many AI workloads need only a fraction of a GPU. Giving each one an entire device can leave capacity idle; sharing GPUs lets platform teams serve more workloads with the infrastructure they already have. NVIDIA GPUs already support several approaches to sharing. Multi-instance GPU (MIG) splits a GPU into hardware-isolated instances that are transparent to the application… Source

NVIDIA RTX Spark is a new Windows PC platform for AI development, creation, and gaming in slim laptops and compact desktops. Powered by an NVIDIA Blackwell RTX... NVIDIA RTX Spark is a new Windows PC platform for AI development, creation, and gaming in slim laptops and compact desktops. Powered by an NVIDIA Blackwell RTX GPU and NVIDIA Grace CPU, it combines up to 128 GB of unified memory with the full NVIDIA CUDA stack and RTX technologies including ray tracing and DLSS 5. This post covers the following technologies for game developers building on RTX… Source

As AI adoption expands globally, people across languages and cultures are using it in more aspects of their daily lives. General-purpose language models serve... As AI adoption expands globally, people across languages and cultures are using it in more aspects of their daily lives. General-purpose language models serve as a powerful foundation, but they sometimes lack the local knowledge, cultural nuance, and precision a particular application needs. Post-training can adapt an open model to these needs. With suitable data and evaluation… Source

Generating tokens faster starts with identifying what limits inference performance. In the batch-size-1 evaluation of Ling-3.0-flash from Inclusion AI on NVIDIA... Generating tokens faster starts with identifying what limits inference performance. In the batch-size-1 evaluation of Ling-3.0-flash from Inclusion AI on NVIDIA DGX Spark, profiling showed that decoding was memory-bound: reading model weights drove execution time. Two operations stood out—the mixture-of-experts (MoE) computation and the output-layer projection that produces scores across the model’… Source

AI Factories are some of the most technologically complex operational environments ever created. Hundreds of thousands of GPUs, NVIDIA Spectrum-X Ethernet... AI Factories are some of the most technologically complex operational environments ever created. Hundreds of thousands of GPUs, NVIDIA Spectrum-X Ethernet scale-out, scale-across and scale-in fabrics, NVIDIA NVLink scale-up fabrics, distributed schedulers, and multi-week training runs operate in continuous lockstep.These layers function as one production system even when different teams own their… Source

Training large-scale AI models on thousands of GPUs is a high-stakes operation where monitoring and automation must be built-in to optimize cluster resiliency... Training large-scale AI models on thousands of GPUs is a high-stakes operation where monitoring and automation must be built-in to optimize cluster resiliency and uptime at scale. This concept isn’t new to data center operators who have measured mean-time-to-repair (MTTR) and mean-time-between-failure (MTBF) long before AI factories existed. Now the stakes are higher. Modern AI factories run… Source

The Global Performance Tuner is a feature for automatic performance tuning released in NVIDIA TensorRT 11.1. It provides automatic, accuracy-aware and... The Global Performance Tuner is a feature for automatic performance tuning released in NVIDIA TensorRT 11.1. It provides automatic, accuracy-aware and end-to-end performance tuning within TensorRT by customizing different optimizations for different models. For more details, see Part 1 of this series, which introduces the challenges associated with scheduling (a critical stage in deep learning… Source

Deploying a trained neural network involves more than translating each operator into a GPU kernel. It is a compilation problem with many legitimate answers. For... Deploying a trained neural network involves more than translating each operator into a GPU kernel. It is a compilation problem with many legitimate answers. For example, when NVIDIA TensorRT builds an engine, it makes interdependent decisions about graph-IR transformations, fusion, implementation selection, memory placement, and quantization. This process is called scheduling. Source

Teaching a robot to navigate an obstacle course requires more than a walking policy. It must perceive its surroundings, maintain balance, and adjust its actions... Teaching a robot to navigate an obstacle course requires more than a walking policy. It must perceive its surroundings, maintain balance, and adjust its actions when obstacles block the path ahead. Developing these capabilities connects several stages: creating assets and an environment, using physics simulation to model how the robot interacts with its environment, and training and evaluating… Source

GPU applications are processing ever-growing datasets with increasingly powerful accelerators. According to the 2025 Stanford AI Index, compute for notable AI... GPU applications are processing ever-growing datasets with increasingly powerful accelerators. According to the 2025 Stanford AI Index, compute for notable AI models has been doubling approximately every five months, while the size of datasets used to train LLMs has been doubling approximately every eight months. As computation and data scale together, moving data efficiently from storage becomes… Source

AI developers aren’t just running a single model anymore—they’re orchestrating multiple models in parallel, chaining inference calls, and experimenting... AI developers aren’t just running a single model anymore—they’re orchestrating multiple models in parallel, chaining inference calls, and experimenting with frontier-class architectures that demand hundreds of gigabytes of memory. NVIDIA DGX Spark was built to bring AI development to the desk, and now you’ll be able to link even more Spark nodes together. At GTC Berlin… Source

Protein design is a search problem. Choose amino acids for an eight-residue protein sequence and more than 25 billion combinations are possible. Choose amino... Protein design is a search problem. Choose amino acids for an eight-residue protein sequence and more than 25 billion combinations are possible. Choose amino acids for a 100-residue protein sequence and the number of possibilities exceeds the number of atoms in the observable universe. Somewhere in that space is a protein that opens a path to a new therapy. There are too many combinations to… Source

An AI agent can complete its assignment and still cross a security boundary. Recent incidents show that agents can circumvent security controls while using... An AI agent can complete its assignment and still cross a security boundary. Recent incidents show that agents can circumvent security controls while using tools, credentials, and services to pursue their assigned goals. Monitoring and enforcement need to operate outside the agent’s reach so that it can’t alter or disable them. NVIDIA Sentry is a security software framework that monitors AI… Source

Autonomous agents can work for hours on complex tasks, accessing files, tools, the internet, and spawning other agents to collaborate with. Executing complex... Autonomous agents can work for hours on complex tasks, accessing files, tools, the internet, and spawning other agents to collaborate with. Executing complex tasks over long time horizons leads to compounding risk: the agents may download code, build new tools, request additional privileges, and touch more of the underlying infrastructure. The agents’ action paths are determined as they work… Source

Building an AI factory digital twin requires developers to bring site plans, equipment assets, and engineering requirements into a consistent model that teams... Building an AI factory digital twin requires developers to bring site plans, equipment assets, and engineering requirements into a consistent model that teams can inspect, revise, and prepare for simulation. Connecting these inputs into an editable workflow helps teams explore design alternatives and incorporate engineering feedback as the design evolves. The NVIDIA Omniverse DSX Blueprint… Source

AI factories turn energy into intelligence. But before an AI workload can produce its first token, operators must bring an interconnected system of compute,... AI factories turn energy into intelligence. But before an AI workload can produce its first token, operators must bring an interconnected system of compute, networking, storage, power, cooling, and software online as one productive unit. In a conventional data center build, teams must first select the site, design power and cooling for the deployment, lay out the data center… Source

NVIDIA CUDA Tile (cuTile) is a Python-based DSL and parallel programming model for NVIDIA GPUs that automatically leverages advanced hardware capabilities to... NVIDIA CUDA Tile (cuTile) is a Python-based DSL and parallel programming model for NVIDIA GPUs that automatically leverages advanced hardware capabilities to deliver peak performance for tile-level GPU programming. By abstracting thread- and warp-level scheduling behind high-level, tile-centric primitives, cuTile enables you to write GPU kernels in Python while compiler optimizations automatically… Source

AI agents are changing how software gets built. Developers have gone from writing every line of code by hand to directing agents that write most of the code. As... AI agents are changing how software gets built. Developers have gone from writing every line of code by hand to directing agents that write most of the code. As a result, the current workflow is split across a laptop, a cloud model API, and a cloud VM. NVIDIA RTX Spark is built for AI development. Thanks to the NVIDIA GB10 system-on-a-chip and large unified memory, developers can run local… Source

Meta launches Muse Spark Open, a coding-optimized 816B mixture-of-experts (MoE) model with 42B active parameters and a 1M token context window. Trained on... Meta launches Muse Spark Open, a coding-optimized 816B mixture-of-experts (MoE) model with 42B active parameters and a 1M token context window. Trained on NVIDIA GB300 NVL72 Blackwell Ultra specifically for agentic coding workflows, it delivers higher first attempt accuracy, reliable tool calling, and deep codebase understanding across repositories. The 1M token context window holds an entire… Source

Always-on agents are taking on longer, more complex enterprise workflows. A single request can involve reading a document, checking information across sources,... Always-on agents are taking on longer, more complex enterprise workflows. A single request can involve reading a document, checking information across sources, following company policy, and calling several tools. NVIDIA Nemotron 3.5 Super VL is an open, customizable multimodal model for these specialized workflows. Its 120B-parameter mixture-of-experts (MoE) architecture activates 12B… Source

AI agents offer remarkable productivity gains in complex chip design workflows, but their tendency to generate plausible yet incorrect artifacts makes... AI agents offer remarkable productivity gains in complex chip design workflows, but their tendency to generate plausible yet incorrect artifacts makes reliability a major barrier to production adoption. In multistage engineering workflows, an undetected error early in setup can silently propagate through downstream stages and final artifact generation. Evaluating and refining these agents against… Source

Enterprise applications have become increasingly demanding. Agentic AI has compounded the need for vast amounts of enterprise data with frequent, complex... Enterprise applications have become increasingly demanding. Agentic AI has compounded the need for vast amounts of enterprise data with frequent, complex queries across increasingly large datasets. Meeting that demand requires greater data throughput without a proportional increase in cost. NVIDIA cuDF addresses this challenge by leveraging the massive parallelism and high-bandwidth memory of… Source

Today, NVIDIA is making the full NVIDIA Run:ai platform available as open source software, giving customers, developers, and ecosystem partners access to... Today, NVIDIA is making the full NVIDIA Run:ai platform available as open source software, giving customers, developers, and ecosystem partners access to production-grade AI workload and GPU orchestration capabilities. This release expands the NVIDIA DSX AI Factory Platform by establishing NVIDIA Run:ai as an open foundation that enables the broader ecosystem to bring differentiated solutions to… Source

Hybrid models interleave state-space model (SSM) layers with a few attention layers. NVIDIA Nemotron 3, Qwen3.5 through Qwen3.8, Kimi-K3, and GLM-5.3-Flash are... Hybrid models interleave state-space model (SSM) layers with a few attention layers. NVIDIA Nemotron 3, Qwen3.5 through Qwen3.8, Kimi-K3, and GLM-5.3-Flash are all built this way, with Mamba-2, Gated DeltaNet (GDN), or Kimi Delta Attention (KDA) as the SSM layer. SSM layers are designed to make decoding fast: they use a fixed-size state in place of a key-value (KV) cache that grows with… Source

Preparing CAD assets for robotics simulation requires more than converting geometry to OpenUSD: developers must configure and validate materials, collision... Preparing CAD assets for robotics simulation requires more than converting geometry to OpenUSD: developers must configure and validate materials, collision geometry, joints, and other physics properties before testing robot behavior. NVIDIA Omniverse libraries, guided by SimReady Foundation specifications and agentic NVIDIA skills, provide a structured workflow for converting and validating… Source

The NVIDIA KGMON team placed second in the KDD Cup 2026 Data Agents competition with a system built around a simple idea of making an agent's harness smaller,... The NVIDIA KGMON team placed second in the KDD Cup 2026 Data Agents competition with a system built around a simple idea of making an agent’s harness smaller, clearer, and easier to verify. The competition asked agents to answer natural-language questions over heterogeneous data sources, including databases, CSV and JSON files, prose documents, PDFs, and briefing videos. Source

How we taught robots to assemble GB300 tester trays and what it taught us about robot learning, mechanical intelligence, and good old-fashioned engineering The... How we taught robots to assemble GB300 tester trays and what it taught us about robot learning, mechanical intelligence, and good old-fashioned engineering The NVIDIA Grace Blackwell GB300 superchip is the engine of AI training and inference for modern foundation models. However, the process of assembling GB300 trays requires skilled physical labor in factories across the world. Source

AI factories are some of the most complex operations in the world, combining GPUs, CPUs, switches, DPUs, and SuperNICs alongside schedulers, orchestration... AI factories are some of the most complex operations in the world, combining GPUs, CPUs, switches, DPUs, and SuperNICs alongside schedulers, orchestration services, security controls, and a rapidly changing software stack. Deploying this infrastructure effectively is a challenge. But so is being able to validate that the infrastructure, software, and policies work together for the workloads the… Source

Supply chain problems are expanding across more SKUs, lanes, and constraints than ever before, while energy grids are balancing more distributed sources in real... Supply chain problems are expanding across more SKUs, lanes, and constraints than ever before, while energy grids are balancing more distributed sources in real time. To meet these challenges, teams need to evaluate larger models and more uncertainty within a practical planning window. NVIDIA cuOpt GPU-accelerated decision optimization can already deliver speedups of more than 10x over CPU… Source

Observatories and telescopes, lasers and X-ray light sources, and other high-throughput instruments generate image data faster than CPU-bound pipelines can... Observatories and telescopes, lasers and X-ray light sources, and other high-throughput instruments generate image data faster than CPU-bound pipelines can process it to support timely decisions. The computational bottleneck is rarely one slow kernel. It’s the whole path from data in the sensor to a decision: read the raw data, match point-spread functions or detector responses, subtract or reduce… Source

Bitwise determinism makes large-scale pretraining easier to debug, validate, and resume reproducibly. These benefits become especially valuable when training... Bitwise determinism makes large-scale pretraining easier to debug, validate, and resume reproducibly. These benefits become especially valuable when training models with trillions of parameters across thousands of GPUs, where multiple parallelism dimensions, low-precision computation, and distributed checkpointing complicate failure reproduction and fix validation. Source

GPU applications increasingly need networking and data movement to behave like first-class GPU-controlled operations rather than host-driven services. When the... GPU applications increasingly need networking and data movement to behave like first-class GPU-controlled operations rather than host-driven services. When the CPU sits in the middle of every network transaction, it becomes a bottleneck on the critical path, adding latency and limiting how efficiently distributed applications can respond in real time. NVIDIA DOCA GPUNetIO… Source

GPU-accelerated Kubernetes clusters depend on compatible versions across dozens of components, each on its own release cycle: host kernels, GPU drivers,... GPU-accelerated Kubernetes clusters depend on compatible versions across dozens of components, each on its own release cycle: host kernels, GPU drivers, container runtimes, networking, storage, operators, and workload frameworks. A configuration that works for one service, GPU generation, and Kubernetes release may silently fail for another, and tracing version conflicts after deployment is… Source

GPU applications increasingly consist of multiple independent components running at the same time within a single process: a latency-sensitive operator... GPU applications increasingly consist of multiple independent components running at the same time within a single process: a latency-sensitive operator alongside a throughput-oriented background kernel; a data preprocessing stage alongside model inference; or multiple stages of a processing workflow sharing a single GPU. Controlling how GPU resources are shared between them remains difficult. Source

AI agents are becoming a standard part of development workflows, but general-purpose agents weren't built with specialized infrastructure software such as... 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. Without domain-specific knowledge, agents may fall back on guesswork. This is an issue in infrastructure development because every correction cycle takes time away from deployment. DOCA is the unified software platform… Source

Adding AI models to local applications requires a portable model format, a reliable runtime, and acceleration that works across target systems. Do Inference Now... Adding AI models to local applications requires a portable model format, a reliable runtime, and acceleration that works across target systems. Do Inference Now (DIN) Deploy is an open-source collection of practical C++ samples that bridges that gap. It combines ONNX Runtime with the NVIDIA TensorRT RTX execution provider to help developers move from a model checkpoint to a native… Source

Automatic speech recognition must handle how people actually speak, not only the languages and styles that dominate pretraining data. Regional dialects and... Automatic speech recognition must handle how people actually speak, not only the languages and styles that dominate pretraining data. Regional dialects and local recording conditions are often underrepresented, so a multilingual model that performs well on broad benchmarks may still fall short in deployment. Saudi Arabic makes that concrete. A model may recognize Modern Standard Arabic or… Source

Generative recommender (GR) systems are emerging as a powerful new direction for large-scale personalization. Instead of treating recommendation as a set of... Generative recommender (GR) systems are emerging as a powerful new direction for large-scale personalization. Instead of treating recommendation as a set of isolated retrieval, ranking, and prediction stages, GRs reformulate recommendation as sequence modeling over user behavior. A user’s interactions, context, candidate items, and actions become tokens in a high-cardinality event stream… Source

AI infrastructure engineers, storage developers, and cloud service providers need fast and secure access to high-capacity file and object storage to support AI... AI infrastructure engineers, storage developers, and cloud service providers need fast and secure access to high-capacity file and object storage to support AI workloads. AI workloads increasingly require high-speed data access for training, fine-tuning, inference context, tool calls, searches, and database lookups. Much of this data lies in files and objects stored both on-premises and in… Source

An agent can finish a task and still take an inefficient path. A failed search can trigger another search. A truncated file read can lead to a command fetching... An agent can finish a task and still take an inefficient path. A failed search can trigger another search. A truncated file read can lead to a command fetching the same content again. A correct final answer hides those extra steps, even though they increase latency and consume tokens. Inefficiencies create more chances for failure. To improve an agent’s behavior… Source

Parallel work, model-family isolation, reversible changes, and GPU-backed validation shaped an open source project designed around coding agents NVIDIA TensorRT... Parallel work, model-family isolation, reversible changes, and GPU-backed validation shaped an open source project designed around coding agents NVIDIA TensorRT Model Connect is an open source collection of AI model reference implementations in C++, built on top of NVIDIA TensorRT. It began with a practical question: could the performance of the NVIDIA inference stack be made accessible to… Source

Vision-language models have made it possible to build visual AI agents that understand video at production scale. The harder problem is turning that capability... Vision-language models have made it possible to build visual AI agents that understand video at production scale. The harder problem is turning that capability into a maintainable system that combines ingestion, stream processing, event detection, retrieval, summarization, and reporting. The NVIDIA Metropolis Blueprint for Video Search and Summarization (VSS) and its agent skills help… Source

To understand where agentic AI stands today, consider the last seismic shift in technology: the rise of the internet in the 90s. It was new and full of... To understand where agentic AI stands today, consider the last seismic shift in technology: the rise of the internet in the 90s. It was new and full of possibilities. You could build a website over a weekend and share it with the world, or chat with someone half way around the world in online chat rooms without long-distance telephone fees. It brought endless opportunity, but also a lot of risk. Source

AI agents can be given a goal, write code, use tools, and keep working as new information becomes available. This opens the door to applications that... AI agents can be given a goal, write code, use tools, and keep working as new information becomes available. This opens the door to applications that investigate software failures, run experiments, and carry out business-critical actions and research over days or weeks. Useful agents need access to workspaces, compute resources, data, credentials, and external services. Source

Every unused watt is capacity left on the table. AI factories are typically provisioned for the unlikely moment when every GPU reaches peak power, creating a... Every unused watt is capacity left on the table. AI factories are typically provisioned for the unlikely moment when every GPU reaches peak power, creating a protective buffer that can leave valuable infrastructure underused during normal operation. NVIDIA DSX MaxLPS uses policy-governed power sharing to allocate power across participating resources dynamically, enabling customers to deploy up to… Source