NVIDIA Dev Blog

In September 2026, NVIDIA announced it is leaning into native GPU programming in Rust. CUDA C++ and CUDA Python are mature, enterprise-grade toolchains, and... In September 2026, NVIDIA announced it is leaning into native GPU programming in Rust. CUDA C++ and CUDA Python are mature, enterprise-grade toolchains, and NVIDIA will be growing and maturing CUDA Rust into 2027 and beyond The systems layer of AI spans inference engines, serving infrastructure, drivers, and agent runtimes, and it churns constantly as models and techniques change. Source

Enterprise work spans messages, decisions, projects, and obligations that change over time. An AI agent that starts without this context must reconstruct it... Enterprise work spans messages, decisions, projects, and obligations that change over time. An AI agent that starts without this context must reconstruct it before contributing. To provide agents with this necessary context, our team used NVIDIA NemoClaw to build a memory-driven Chief of Staff. It maintains a human-readable knowledge layer called the self model: an agent memory of relevant… Source

Running reasoning and agentic AI at the edge has been harder than it needs to be. Until recently, models capable of multi-step reasoning were too large to run... Running reasoning and agentic AI at the edge has been harder than it needs to be. Until recently, models capable of multi-step reasoning were too large to run locally on edge hardware. Developers building agents have had to route inference through a data center, adding network dependency, increasing costs, and exposing data that may need to stay on device. That constraint is lifting. Source

Modern AI platforms are no longer a single application behind one login screen. A user may start in a central portal, open a governed dataset, launch a notebook... Modern AI platforms are no longer a single application behind one login screen. A user may start in a central portal, open a governed dataset, launch a notebook where that data resides, and invoke an assistant that calls services in another cluster. The workflow feels unified, but identity crosses control-plane and data-plane boundaries at every step. That is where conventional single sign-on… Source

AI agents are learning to do more by working together. A lead agent can break a complex task into smaller jobs and assign those jobs to specialized subagents.... AI agents are learning to do more by working together. A lead agent can break a complex task into smaller jobs and assign those jobs to specialized subagents. Additionally, users are starting to run multiple agent sessions at the same time. Multi-agent workflows for accomplishing complex tasks are also becoming more common. This breadth-first approach can improve the speed of task completion… Source

NVIDIA CUDA remains the foundation of GPU-accelerated computing, powering everything from scientific simulations to large-scale AI training. But writing... Source

This post is the third in a series on AI model co-design. It explores how to accelerate LLM inference while maintaining accuracy using speculative decoding and... This post is the third in a series on AI model co-design. It explores how to accelerate LLM inference while maintaining accuracy using speculative decoding and offers five guidelines for selecting draft length and draft mechanism across the Pareto frontier. For a discussion of how model design choices impact both throughput and interactivity without sacrificing accuracy, see AI Model Co… Source

AI is changing the pace of cybersecurity. Agentic systems can coordinate work and pursue complex objectives over long horizons. Security teams are beginning to... AI is changing the pace of cybersecurity. Agentic systems can coordinate work and pursue complex objectives over long horizons. Security teams are beginning to apply agents across security operations, but many implementations remain anchored to existing alerts, predefined workflows, and known attack behaviors. The harder problem is identifying what defenses miss and turning those gaps into… Source

The surge in AI adoption is transforming everything from chatbots to content generation. Still, a common pain point remains: How can organizations confidently... The surge in AI adoption is transforming everything from chatbots to content generation. Still, a common pain point remains: How can organizations confidently size GPU resources for inference workloads and optimize Total Cost of Ownership (TCO)? With a dizzying mix of latency targets, model choices, quirky traffic patterns, and budget constraints, it’s easy to feel lost in the weeds… Source

Agentic AI is changing how research is done. AI scientists can read papers, propose hypotheses, call models, and determine which experiments to prioritize next.... Agentic AI is changing how research is done. AI scientists can read papers, propose hypotheses, call models, and determine which experiments to prioritize next. First proving their value in software engineering, coding agents now write, test, and ship production code. Scientific research can be more demanding and iterative. Researchers continually evaluate evidence, refine hypotheses… Source

A perception stack is shaped by the vehicle that carries it. Move the same software to a new carline—for example, from an SUV to a sedan or another vehicle... A perception stack is shaped by the vehicle that carries it. Move the same software to a new carline—for example, from an SUV to a sedan or another vehicle variant in the portfolio—and its perception of the world changes. The sensor placement, calibration, fields of view, occlusions, body geometry, timing, and coverage all shift. A traffic light may appear in a different part of the frame. Source

Open AI models are evolving faster than ever, but bringing them into native applications can still require model-specific conversion, preprocessing,... Open AI models are evolving faster than ever, but bringing them into native applications can still require model-specific conversion, preprocessing, post-processing, and runtime code. NVIDIA TensorRT Model Connect open collection of reference implementations helps to address this challenge. TensorRT Model Connect shows you how to run supported models with NVIDIA TensorRT in native C++… Source

AI factories must support increasingly large models and more complex reasoning workloads. To keep up with the insatiable compute demands of AI workloads,... AI factories must support increasingly large models and more complex reasoning workloads. To keep up with the insatiable compute demands of AI workloads, hyperscalers and AI-native companies are developing custom AI accelerators, or XPUs. Deploying these accelerators at scale requires high-bandwidth memory (HBM) to keep compute fed, sufficient package and silicon area for more compute… Source

Navigation enables a robot to turn perception and motion into purposeful autonomy. Unlike locomotion, which produces stable movement, navigation must be used to... Navigation enables a robot to turn perception and motion into purposeful autonomy. Unlike locomotion, which produces stable movement, navigation must be used to continuously localize the robot, interpret changing surroundings, select a route, and avoid obstacles to reach a goal safely. Moving this capability to a new robot or scene can require new data, simulation assets, robot interfaces… Source

Alibaba released the model weights for Qwen3.8-Flash-Next as a preview of the upcoming Qwen4 architecture for developers to experiment with and evaluate. It’s... Alibaba released the model weights for Qwen3.8-Flash-Next as a preview of the upcoming Qwen4 architecture for developers to experiment with and evaluate. It’s a multimodal mixture-of-experts (MoE) model with a 125B-parameter main model supplemented by an additional 51B N-gram embeddings, with 6B parameters activated per token. It has a native 262,144-token context window, extensible to 1M tokens… Source

When an LLM engine process fails, the standard recovery path involves a cold restart. This requires loading weights into HBM from storage, compiling kernels,... When an LLM engine process fails, the standard recovery path involves a cold restart. This requires loading weights into HBM from storage, compiling kernels, and capturing NVIDIA CUDA graphs. For large models, initialization can take several minutes, during which surviving workers must absorb the displaced traffic. Shadow engine recovery, available as a preview feature in NVIDIA Dynamo… Source

For years, a Python developer who needed a GPU had two realistic choices: Learn NVIDIA CUDA C++ well enough to write an extension, set up a build toolchain, and... For years, a Python developer who needed a GPU had two realistic choices: Learn NVIDIA CUDA C++ well enough to write an extension, set up a build toolchain, and maintain bindings back to Python, which most people never did; or move up the stack and let someone else’s library do it, namely PyTorch, CuPy, or RAPIDS. The second option is why the Python GPU ecosystem thrives. But it has limits. Source

The massive growth of generative AI has fundamentally altered data center design. As distributed model training scales to span hundreds of thousands of GPUs,... The massive growth of generative AI has fundamentally altered data center design. As distributed model training scales to span hundreds of thousands of GPUs, the scale-out network connecting these nodes has emerged as a first-order performance bottleneck. For decades, traditional off-the-shelf Ethernet has been the undisputed king of enterprise and cloud networking. It is cheap, standardized… Source

AI agents have expanded inference from single-turn interactions into multi-step workflows that reason, invoke tools, coordinate subagents, and carry growing... AI agents have expanded inference from single-turn interactions into multi-step workflows that reason, invoke tools, coordinate subagents, and carry growing context from one turn to the next. The scale of this shift is now visible in raw consumption: across 100 trillion tokens of real-world usage, OpenRouter’s State of AI report found that average prompt tokens per request grew roughly fourfold… Source

Traditional cloud infrastructure was designed for predictable, general-purpose workloads and standard interfaces. Agentic AI factories connect diverse users,... Traditional cloud infrastructure was designed for predictable, general-purpose workloads and standard interfaces. Agentic AI factories connect diverse users, agents, applications, data sources, and storage systems to massively accelerated compute at multi-terabit bandwidth per server, making dedicated DPU processing essential for line-rate networking, storage, and security. Source

AI factories are interconnected systems where fleet economics depend on how efficiently the entire stack converts power and capital into completed agent tasks.... AI factories are interconnected systems where fleet economics depend on how efficiently the entire stack converts power and capital into completed agent tasks. While GPUs run the models, CPUs handle orchestration, tool execution, and sandboxed computation. Unlike conventional computing with stable runtime profiles, agentic workloads are unpredictable and highly variable. Based on telemetry from… Source

NVIDIA Groq 3 LPX is the interactive AI inference accelerator for the NVIDIA Vera Rubin platform. At the core of the platform is NVIDIA Vera Rubin NVL72, the... NVIDIA Groq 3 LPX is the interactive AI inference accelerator for the NVIDIA Vera Rubin platform. At the core of the platform is 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. Groq 3 LPX, when paired with Vera Rubin NVL72, extends the platform’s… Source

Use AdaptGrow, a GPU-accelerated matrix factorization algorithm, to turn rolling correlation and tail-dependence matrices into hard clusters, soft factor... Use AdaptGrow, a GPU-accelerated matrix factorization algorithm, to turn rolling correlation and tail-dependence matrices into hard clusters, soft factor loadings, and structural-break signals at single-GPU and multi-node scale Quant strategies routinely group instruments for portfolio construction, risk aggregation, statistical arbitrage, and trade surveillance. Incorrect groupings can make… Source

AI factories are power-constrained industrial systems. The question is no longer how many GPUs fit in a data center, but how much AI output each available... AI factories are power-constrained industrial systems. The question is no longer how many GPUs fit in a data center, but how much AI output each available megawatt can deliver. For AI inference workloads, this makes application-level performance per watt the key metric for measuring AI factory efficiency. Not every megawatt translates to revenue-generating compute. Power distribution, cooling… Source

A frontier language model is only one component of an AI agent. The surrounding agent system—often called a harness—determines how the model receives... A frontier language model is only one component of an AI agent. The surrounding agent system—often called a harness—determines how the model receives context, uses tools, maintains state, responds to feedback, recovers from failure, and sustains progress over long-running tasks. The challenge is how to build the agent architecture that makes frontier language models work reliably on extended… Source

As AI agents become more capable and operate over longer horizons, building security and trust into the applications they power becomes increasingly important.... As AI agents become more capable and operate over longer horizons, building security and trust into the applications they power becomes increasingly important. Drawing on work with NVIDIA OpenShell, agent developers, open-source projects, and partners across the ecosystem, AI safety and security teams at NVIDIA offer their perspective on the emerging agent stack—including the role of each layer… Source

Recommender systems (RecSys) are one of the most ubiquitous machine learning problems in the consumer internet industry yet notoriously difficult to train and... Recommender systems (RecSys) are one of the most ubiquitous machine learning problems in the consumer internet industry yet notoriously difficult to train and serve at scale. The advent of LLMs has inspired a shift from the traditional embedding-similarity-based objective to a generative one, where the goal is to predict the next action or item in a large catalog given a sequence of user histories. Source

NVIDIA Holoscan is a platform for building real-time AI applications at the edge, from medical imaging to robotics. HoloHub is its companion repository: a... NVIDIA Holoscan is a platform for building real-time AI applications at the edge, from medical imaging to robotics. HoloHub is its companion repository: a growing collection of reference applications and components that demonstrate what’s possible. We wanted to explore how a general-purpose coding agent could use the same examples, documentation, and development tools available to an engineer… Source

Modern vision-language models (VLMs) can support tasks such as visual question answering, captioning, and image-text reasoning. In practice, however, the data... Modern vision-language models (VLMs) can support tasks such as visual question answering, captioning, and image-text reasoning. In practice, however, the data needed to adapt these models may be distributed across institutions or organizations that cannot centralize their raw records. Federated learning provides a way to coordinate training across these data-local sites. For VLMs… Source

Robots need policies that can adapt to their sensors, environments, and tasks while running on onboard computing hardware. World models offer a foundation for... Robots need policies that can adapt to their sensors, environments, and tasks while running on onboard computing hardware. World models offer a foundation for learning physical interactions, but their size can make on-device deployment difficult. This changes with the new NVIDIA Cosmos 3 Edge. Cosmos 3 Edge is a 4B omni-model (with a 2B NVIDIA Nemotron-based reasoner) in the Cosmos 3 family. Source

AI agents are only as effective as the context they receive. Even with capable models and well-documented NVIDIA libraries, agents can spend extra steps finding... AI agents are only as effective as the context they receive. Even with capable models and well-documented NVIDIA libraries, agents can spend extra steps finding the right tools, burn tokens on dead ends, or struggle with specialized tasks. Skills package the instructions, examples, and tool guidance for agents to move faster from intent to solution. To measure whether these skills improve agent… Source

Atomistic simulation requires three things: knowledge of the science, compute-efficient implementation of simulations, and accessible interfaces to the... Atomistic simulation requires three things: knowledge of the science, compute-efficient implementation of simulations, and accessible interfaces to the simulation stack. The first remains the researcher’s domain, as no tool substitutes for knowing what to simulate or recognizing a physically meaningful result. NVIDIA ALCHEMI Toolkit, introduced earlier this year, has dramatically reduced the… Source

Uniform Manifold Approximation and Projection (UMAP) is a dimensionality reduction technique widely used for visualization and feature extraction. Applications... Uniform Manifold Approximation and Projection (UMAP) is a dimensionality reduction technique widely used for visualization and feature extraction. Applications range across exploratory data analysis, topic modeling, and single-cell analysis. Many of these workflows are iterative and exploratory, requiring UMAP to be run repeatedly as users analyze their data or tune parameters. As datasets grow… Source

Teams customize their models to hit their targets for latency, speed, memory, and compute. With the open NVIDIA Nemotron family of models, developers can find... Teams customize their models to hit their targets for latency, speed, memory, and compute. With the open NVIDIA Nemotron family of models, developers can find the right-sized model for their needs. The new Nemotron 3.5 Lightning NVFP4 checkpoint, for example, preserves accuracy while unlocking up to 4x faster throughput. It’s compressed down to 22 GB from the 66 GB full precision checkpoint… Source

Alibaba released the open weights for Qwen3.8-2.4T-A95B (Qwen3.8-Max), its largest open-weight model, bringing near-frontier capabilities to the open... Alibaba released the open weights for Qwen3.8-2.4T-A95B (Qwen3.8-Max), its largest open-weight model, bringing near-frontier capabilities to the open ecosystem. It has 2.4T total parameters with 95B activated per token. It has 2.4T total parameters with 95B activated per token. It’s a fine-grained mixture of experts (MoE) architecture with a hybrid of full and linear attention, a context window of… Source

AI infrastructure spans multiple layers, from compute and networking to storage, orchestration, and applications. When performance degrades, identifying the... AI infrastructure spans multiple layers, from compute and networking to storage, orchestration, and applications. When performance degrades, identifying the source can be difficult because a symptom observed at one layer may originate elsewhere in the stack. A full-stack observability strategy connects telemetry across these layers, helping infrastructure and operations teams detect problems… Source

Video is a core data path across NVIDIA Jetson applications, from robotics and intelligent video analytics to industrial automation, healthcare, media... Video is a core data path across NVIDIA Jetson applications, from robotics and intelligent video analytics to industrial automation, healthcare, media processing, and remote operations. A system may capture several cameras, decode network streams, run AI inference or conventional vision processing, draw results, and encode video for storage or delivery. The individual calls are… Source

Long-running AI agents spend most of their time on high-volume execution: tool calls, result validation, and subagent delegation. Using a frontier reasoning... Long-running AI agents spend most of their time on high-volume execution: tool calls, result validation, and subagent delegation. Using a frontier reasoning model for every execution step adds cost and latency. NVIDIA Nemotron 3.5 Lightning is an open 30B mixture-of-experts (MoE) model with 3B active parameters built for that execution layer of always-on agents. It is designed for harnesses… Source

Building an AI agent does not end with choosing a single model. Each model has its own strengths, weaknesses, and cost profile, which can shift from one... Building an AI agent does not end with choosing a single model. Each model has its own strengths, weaknesses, and cost profile, which can shift from one workload to another—or even within the same workload. For example, an agentic task may need classification for one step, reasoning for the next, and a smaller model for routine follow-up tasks. Sending every request to the largest model can… Source

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... 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. Optimized to run across a range of NVIDIA edge, desktop, and workstation AI platforms, Muse Glimmer delivers 20K tokens/sec on a single GPU, enabling always-on agents to process data locally and execute complex… Source

A central challenge in robotics is building policies that generalize beyond the demonstrations they’re trained on. A policy that succeeds in a training scene... A central challenge in robotics is building policies that generalize beyond the demonstrations they’re trained on. A policy that succeeds in a training scene often fails when object shapes, positions, or lighting change. Generalizing to these new conditions requires the policy to understand the tasks underlying physics, not just mimic the demonstrations. This ability comes from the backbone it’s… Source

Autonomous vehicle (AV) development often relies on separate models for trajectory generation, high-level intent prediction, scene understanding, and data... Autonomous vehicle (AV) development often relies on separate models for trajectory generation, high-level intent prediction, scene understanding, and data labeling. This separation makes it hard to compare related outputs, investigate model behavior, and reuse the same representations across the development workflow. NVIDIA Alpamayo 2 Super is an open 34-billion-parameter reasoning vision… Source

Running a dedicated Kubernetes cluster per team often results in more isolation than an organization requires. While one cluster can be successfully shared... Running a dedicated Kubernetes cluster per team often results in more isolation than an organization requires. While one cluster can be successfully shared across many teams, the coordination costs increase as the number of teams grows. Challenges include conflicting CRD versions, overlapping RBAC, and no clean way to carve GPU capacity into team-level budgets. At a certain scale… Source

Storage is an active part of every agentic AI workflow. As agents retrieve enterprise knowledge, access persistent memory, reuse key-value (KV) cache data,... Storage is an active part of every agentic AI workflow. As agents retrieve enterprise knowledge, access persistent memory, reuse key-value (KV) cache data, execute tools, and generate new results, storage systems must continuously supply and preserve the data that moves the agent reasoning loop. Each agent step can trigger multiple storage operations, and those operations can repeat across… Source

As agentic and long-context workloads become common, the context lengths increase and attention consumes a larger share of inference time (Figure 1). Because... As agentic and long-context workloads become common, the context lengths increase and attention consumes a larger share of inference time (Figure 1). Because attention now dominates that cost, how it is designed—not just how it is implemented—increasingly determines a model’s inference performance. Shaping model architecture around how GPUs execute it is the premise of AI model co-design. Source

The demand for high-quality video continues to accelerate across industries, powering everything from immersive streaming experiences to remote collaboration,... The demand for high-quality video continues to accelerate across industries, powering everything from immersive streaming experiences to remote collaboration, generative AI media tools, and large-scale content delivery. Behind these experiences is a growing need for video pipelines that are faster, more efficient, and capable of handling increasingly complex formats and workloads. Source

NVIDIA nvmath-python is a library designed to bridge the gap between the Python scientific community and NVIDIA CUDA-X math libraries. It gives Python users... NVIDIA nvmath-python is a library designed to bridge the gap between the Python scientific community and NVIDIA CUDA-X math libraries. It gives Python users access to CUDA-X performance for common math operations without disrupting existing workflows. Depending on the API, operations can run on a CPU, CUDA-enabled GPU, or distributed multi-GPU, multi-node systems. Source

Knowledge workers are increasingly integrating AI agents into their workflows. Agents that function as "digital coworkers" offer clear benefits. For example,... Knowledge workers are increasingly integrating AI agents into their workflows. Agents that function as “digital coworkers” offer clear benefits. For example, they can review a bug report, implement and test a fix, push a patch, and ping a human for review. By handling routine tasks, agents have the potential to deliver large productivity gains. On the other hand, connecting a large language model… Source

Two AI computing clusters built from identical NVIDIA H100, GB200 NVL72, or GB300 NVL72 systems can deliver materially different training throughput. We... Two AI computing clusters built from identical NVIDIA H100, GB200 NVL72, or GB300 NVL72 systems can deliver materially different training throughput. We routinely see 8% to 12% gaps between partner deployments and the corresponding NVIDIA reference architecture (RA) on the same workload, same model, same global batch size. The cause is often a stack of configuration choices in the kernel… Source

Deploying an AI coding assistant in a regulated, sovereign, or source-sensitive environment, often comes with challenges. Three common issues are: the source... Deploying an AI coding assistant in a regulated, sovereign, or source-sensitive environment, often comes with challenges. Three common issues are: the source cannot leave the network, the assistant occasionally invents package names that introduce supply-chain risk, and there is no audit trail when a generated change ships a defect. This tutorial walks you through how to self-host a validated… Source