NVIDIA Technical Blog

Recent

See all
Jul 07, 2025

Think Smart and Ask an Encyclopedia-Sized Question: Multi-Million Token Real-Time Inference for 32X More Users

Modern AI applications increasingly rely on models that combine huge parameter counts with multi-million-token context windows. Whether it is AI agents...
8 MIN READ
Decorative image.
Jul 07, 2025

NVIDIA cuQuantum Adds Dynamic Gradients, DMRG, and Simulation Speedup 

NVIDIA cuQuantum is an SDK of optimized libraries and tools that accelerate quantum computing emulations at both the circuit and device level by orders of...
5 MIN READ
Jul 07, 2025

Turbocharging AI Factories with DPU-Accelerated Service Proxy for Kubernetes

As AI evolves to planning, research, and reasoning with agentic AI, workflows are becoming increasingly complex. To deploy agentic AI applications efficiently,...
6 MIN READ
Jul 07, 2025

LLM Inference Benchmarking: Performance Tuning with TensorRT-LLM

This is the third post in the large language model latency-throughput benchmarking series, which aims to instruct developers on how to benchmark LLM inference...
11 MIN READ
Jul 03, 2025

RAPIDS Adds GPU Polars Streaming, a Unified GNN API, and Zero-Code ML Speedups

RAPIDS, a suite of NVIDIA CUDA-X libraries for Python data science, released version 25.06, introducing exciting new features. These include a Polars GPU...
6 MIN READ
Jul 03, 2025

New Video: Build Self-Improving AI Agents with the NVIDIA Data Flywheel Blueprint

AI agents powered by large language models are transforming enterprise workflows, but high inference costs and latency can limit their scalability and user...
2 MIN READ
Jul 02, 2025

Advanced NVIDIA CUDA Kernel Optimization Techniques: Handwritten PTX

As accelerated computing continues to drive application performance in all areas of AI and scientific computing, there's a renewed interest in GPU optimization...
11 MIN READ
Jul 02, 2025

NVIDIA Omniverse: What Developers Need to Know About Migration Away From Launcher

As part of continued efforts to ensure NVIDIA Omniverse is a developer-first platform, NVIDIA will be deprecating the Omniverse Launcher on Oct. 1. Doing so...
2 MIN READ

Inference Performance

See all
Decorative image.
Jul 01, 2025

Per-Tensor and Per-Block Scaling Strategies for Effective FP8 Training

In this blog post, we’ll break down the main FP8 scaling strategies—per-tensor scaling, delayed and current scaling, and per-block scaling (including the...
10 MIN READ
Jun 26, 2025

Run Google DeepMind’s Gemma 3n on NVIDIA Jetson and RTX

As of today, NVIDIA now supports the general availability of Gemma 3n on NVIDIA RTX and Jetson. Gemma, previewed by Google DeepMind at Google I/O last month,...
4 MIN READ
Jun 24, 2025

Introducing NVFP4 for Efficient and Accurate Low-Precision Inference

To get the most out of AI, optimizations are critical. When developers think about optimizing AI models for inference, model compression techniques—such as...
11 MIN READ
Decorative image.
Jun 13, 2025

Run High-Performance LLM Inference Kernels from NVIDIA Using FlashInfer​​

Best-in-class LLM Inference requires two key elements: speed and developer velocity. Speed refers to maximizing the efficiency of the underlying hardware by...
6 MIN READ
A decorative image.
Jun 12, 2025

Run High-Performance AI Applications with NVIDIA TensorRT for RTX

NVIDIA TensorRT for RTX is now available for download as an SDK that can be integrated into C++ and Python applications for both Windows and Linux. At...
7 MIN READ
Jun 06, 2025

How NVIDIA GB200 NVL72 and NVIDIA Dynamo Boost Inference Performance for MoE Models

The latest wave of open source large language models (LLMs), like DeepSeek R1, Llama 4, and Qwen3, have embraced Mixture of Experts (MoE) architectures. Unlike...
12 MIN READ
May 22, 2025

Blackwell Breaks the 1,000 TPS/User Barrier With Meta’s Llama 4 Maverick

NVIDIA has achieved a world-record large language model (LLM) inference speed. A single NVIDIA DGX B200 node with eight NVIDIA Blackwell GPUs can achieve over...
9 MIN READ
May 21, 2025

NVIDIA Dynamo Accelerates llm-d Community Initiatives for Advancing Large-Scale Distributed Inference

The introduction of the llm-d community at Red Hat Summit 2025 marks a significant step forward in accelerating generative AI inference innovation for the open...
5 MIN READ

Generative AI

See all
Jul 02, 2025

Optimizing FLUX.1 Kontext for Image Editing with Low-Precision Quantization

FLUX.1 Kontext, the recently released model from Black Forest Labs, is a fascinating addition to the repertoire of community image generation models. The open...
10 MIN READ
Jul 01, 2025

How to Build Custom AI Agents with NVIDIA NeMo Agent Toolkit Open Source Library

AI agents are revolutionizing the digital workforce by transforming business operations, automating complex tasks, and unlocking new efficiencies. With the...
3 MIN READ
Jun 30, 2025

Best-in-Class Multimodal RAG: How the Llama 3.2 NeMo Retriever Embedding Model Boosts Pipeline Accuracy

Data goes far beyond text—it is inherently multimodal, encompassing images, video, audio, and more, often in complex and unstructured formats. While the...
7 MIN READ
Jun 30, 2025

NVIDIA NeMo Retriever Scores First Place for Visual Retrieval

NeMo Retriever tops several visual document retrieval leaderboards, setting new standards for RAG apps.
1 MIN READ
Jun 25, 2025

Check Out Sovereign AI in Practice Through an NVIDIA Webinar

Join NVIDIA experts and leading European model builders on July 8 for a webinar on building and deploying multilingual large language models.
1 MIN READ
Jun 25, 2025

How to Streamline Complex LLM Workflows Using NVIDIA NeMo-Skills

A typical recipe for improving LLMs involves multiple stages: synthetic data generation (SDG), model training through supervised fine-tuning (SFT) or...
10 MIN READ
Jun 25, 2025

Boost Embedding Model Accuracy for Custom Information Retrieval

Customizing embedding models is crucial for effective information retrieval, especially when working with domain-specific data like legal text, medical records,...
8 MIN READ
Jun 24, 2025

NVIDIA Run:ai and Amazon SageMaker HyperPod: Working Together to Manage Complex AI Training

NVIDIA Run:ai and Amazon Web Services have introduced an integration that lets developers seamlessly scale and manage complex AI training workloads. Combining...
5 MIN READ

Data Science

See all
Jun 27, 2025

How to Work with Data Exceeding VRAM in the Polars GPU Engine

In high-stakes fields such as quant finance, algorithmic trading, and fraud detection, data practitioners frequently need to process hundreds of gigabytes (GB)...
4 MIN READ
Jun 27, 2025

AI Analyzes Nurses’ Observations to Reduce Patient Danger

Researchers have developed an AI-powered tool that can analyze nurses’ shift notes to identify—far earlier than traditional methods—when an admitted...
4 MIN READ
Jun 24, 2025

Upcoming Livestream: Beyond the Algorithm With NVIDIA

Join us on June 26 to learn how to distill cost-efficient models with the NVIDIA Data Flywheel Blueprint.
1 MIN READ
Jun 18, 2025

AI in Manufacturing and Operations at NVIDIA: Accelerating ML Models with NVIDIA CUDA-X Data Science

NVIDIA leverages data science and machine learning to optimize chip manufacturing and operations workflows—from wafer fabrication and circuit probing to...
8 MIN READ
Jun 16, 2025

AI Aims to Bring Order to the Law

A team of Stanford University researchers has developed an LLM system to cut through bureaucratic red tape. The LLM—dubbed the System for Statutory Research,...
4 MIN READ
Jun 13, 2025

New Professional Certifications in Accelerated Data Science & AI Networking

Unlock your potential with the new NCP-Accelerated Data Science and AI Networking certifications. Validate your skills in GPU-accelerated tools, data science...
1 MIN READ
Jun 13, 2025

Live Webinar: What’s New With NVIDIA Certification

Join this multi-time zone webinar on learning more about the NVIDIA Certifications. Learn the practical prep tips from NVIDIA Certification experts, insights on...
1 MIN READ
Jun 12, 2025

Accelerated Sequence Alignment for Protein Science with MMseqs2-GPU and NVIDIA NIM

Protein sequence alignment—comparing protein sequences for similarities—is fundamental to modern biology and medicine. It illuminates gene functions by...
9 MIN READ

Robotics

See all
Jun 24, 2025

Making Industrial Robots More Nimble With NVIDIA Isaac Manipulator and Vention MachineMotion AI

As industrial automation accelerates, factories are increasingly relying on advanced robotics to boost productivity and operational resilience. The successful...
7 MIN READ
Jun 17, 2025

R²D²: Building AI-based 3D Robot Perception and Mapping with NVIDIA Research

Robots must perceive and interpret their 3D environments to act safely and effectively. This is especially critical for tasks such as autonomous navigation,...
13 MIN READ
Jun 16, 2025

Isaac Sim and Isaac Lab Are Now Available for Early Developer Preview

NVIDIA today released developer previews of NVIDIA Isaac Sim and NVIDIA Isaac Lab — reference robotics simulation and learning frameworks. Now available on...
5 MIN READ
Jun 16, 2025

Enhance Robot Learning with Synthetic Trajectory Data Generated by World Foundation Models

Generalist robotics have arrived, powered by advances in mechatronics and robot AI foundation models. But a key bottleneck remains: robots need vast training...
8 MIN READ
Decorative image.
Jun 12, 2025

NVIDIA Holoscan Sensor Bridge Empowers Developers with Real-Time Data Processing

In the rapidly evolving robotics and edge AI landscape, the ability to efficiently process and transfer sensor data is crucial. Many edge applications are...
9 MIN READ
Decorative image.
Jun 11, 2025

Scale Realistic Robot Simulation Using the NVIDIA NeMo Agent Toolkit for Physical AI

Physical AI enables autonomous systems—think robots, self-driving cars, and smart spaces—to perceive, understand, and act intelligently in the real world....
10 MIN READ
Jun 11, 2025

Develop Custom Physical AI Foundation Models with NVIDIA Cosmos Predict-2

Building smarter robots and autonomous vehicles (AVs) starts with physical AI models that understand real-world dynamics. These models serve two critical roles:...
7 MIN READ
May 30, 2025

How Robot Brains Dream and Explore Unseen Worlds

NVIDIA Isaac GR00T-Dreams enables developers to generate large-scale synthetic trajectory data from minimal human demonstrations, enabling robots to quickly...
1 MIN READ

Simulation / Modeling / Design

See all
Jun 27, 2025

Just Released: NVIDIA PhysicsNeMo v25.06

New functionality to curate and train DoMINO at scale and validate against a physics-based benchmark suite.
1 MIN READ
Jun 18, 2025

Compiler Explorer: An Essential Kernel Playground for CUDA Developers

Have you ever wondered exactly what the CUDA compiler generates when you write GPU kernels? Ever wanted to share a minimal CUDA example with a colleague...
7 MIN READ
Decorative image.
Jun 12, 2025

Streamlining GPU Porting for EDF's Fluid Dynamics Simulations with NVIDIA Nsight Profilers

Porting existing CPU applications to NVIDIA GPUs can unlock performance gains, enabling users to solve problems at a much greater scale and speed. While the...
6 MIN READ
cuEquivariance expands to accelerate next-gen protein structure models
Jun 11, 2025

Accelerated Molecular Modeling with NVIDIA cuEquivariance and NVIDIA NIM microservices

The emergence of models like AlphaFold2 has skyrocketed the demand for faster inference and training of molecular AI models. The need for speed comes with...
8 MIN READ
Jun 11, 2025

Building Photorealistic Digital Twins With Siemens Teamcenter Digital Reality Viewer

Modern products often consist of millions of parts and require intricate design and collaboration. The industrial world is facing significant challenges in...
4 MIN READ
A GIF showing autonomous vehicles.
Jun 11, 2025

Simplify End-to-End Autonomous Vehicle Development with New NVIDIA Cosmos World Foundation Models

The shift to end-to-end planning models for powering autonomous vehicles (AVs) is increasing the demand for high-quality, physically-based sensor data. These...
7 MIN READ
Jun 11, 2025

Accelerating AV Simulation with Neural Reconstruction and World Foundation Models

Autonomous vehicle (AV) stacks are evolving from a hierarchy of discrete building blocks to end-to-end architectures built on foundation models. This transition...
7 MIN READ
Jun 10, 2025

Transforming Quantum Education with AI Supercomputing and NVIDIA CUDA-Q Academic

As quantum computers scale, they will integrate with AI supercomputers to tackle some of the world’s most challenging problems. These accelerated quantum...
8 MIN READ

Computer Vision / Video Analytics

See all
A decorative image.
Jun 18, 2025

Run Multimodal Extraction for More Efficient AI Pipelines Using One GPU

As enterprises generate and consume increasing volumes of diverse data, extracting insights from multimodal documents, like PDFs and presentations, has become a...
8 MIN READ
Jun 08, 2025

AI Helps Locate Dangerous Fishing Nets Lost at Sea

Conservationists have launched a new AI tool that can sift through petabytes of underwater imaging from anywhere in the world to identify signs of abandoned or...
4 MIN READ
A decorative image.
May 23, 2025

Unlock Efficient Data Processing with the Latest from NVIDIA DALI

NVIDIA DALI, a portable, open source software library for decoding and augmenting images, videos, and speech, recently introduced several features that improve...
8 MIN READ
May 18, 2025

Advance Video Analytics AI Agents Using the NVIDIA AI Blueprint for Video Search and Summarization

Vision language models (VLMs) have transformed video analytics by enabling broader perception and richer contextual understanding compared to traditional...
15 MIN READ
May 08, 2025

Revolutionizing Neural Reconstruction and Rendering in gsplat with 3DGUT

Realistic 3D simulation is becoming a cornerstone of modern AI and graphics, from training autonomous vehicles (AV) to powering robotics and digital twins....
5 MIN READ
Apr 24, 2025

Benchmarking Agentic LLM and VLM Reasoning for Gaming with NVIDIA NIM

This is the first post in the LLM Benchmarking series, which shows how to use GenAI-Perf to benchmark the Meta Llama 3 model when deployed with NVIDIA NIM. ...
7 MIN READ
A heatmap animated GIF.
Apr 16, 2025

AI-Generated Heat Maps Keep Seniors and their Privacy Safe

By 2030, more than one in five Americans will be 65 or older, becoming the United States’ largest group of seniors ever. Silicon Valley-based startup Butlr...
4 MIN READ
Apr 11, 2025

AI Advances Parkinson’s Detection Using Standard MRI Scans

A simple brain scan may soon be all that's needed to accurately diagnose Parkinson’s disease, thanks to a new AI-powered tool. The advancement could help...
3 MIN READ

Content Creation / Rendering

See all
banner for the Project G-Assist Hackathon
Jun 17, 2025

Getting Started with Project G-Assist: Build a Twitch-Integrated Plug-in

Today, tweaking your PC to suit your workflows often involves digging through menus and settings across multiple control panels. Project G-Assist is an...
7 MIN READ
Jun 13, 2025

ICYMI: NVIDIA RTX PRO AI Workstations Enable AI-Powered Podcast Creation

Transform your PDFs into personalized audio using NVIDIA RTX PRO and the PDF to Podcast AI Blueprint.
1 MIN READ
Jun 05, 2025

Vortex Delivers CT-Like Ultrasound to Doctors Offices With NVIDIA Jetson

Despite advances in medical imaging, many medical professionals still lack access to diagnostic imaging in their own offices. Vortex Imaging—a medical imaging...
7 MIN READ
Jun 02, 2025

NVIDIA Releases RTX Neural Rendering Tech for Unreal Engine Developers

Artificial intelligence is bridging the gap between game visuals and state-of-the-art CGI in films. It is evolving traditional graphics programming and giving...
5 MIN READ
A still from the game, Indiana Jones and the Great Circle.
May 15, 2025

Path Tracing Optimizations in Indiana Jones™: Opacity MicroMaps and Compaction of Dynamic BLASs

The first post in this series, Path Tracing Optimization in Indiana Jones™: Shader Execution Reordering and Live State Reductions, covered ray-gen shader...
13 MIN READ
May 15, 2025

Path Tracing Optimization in Indiana Jones™: Shader Execution Reordering and Live State Reductions

This post is part of the Path Tracing Optimizations in Indiana Jones™ series.   While adding a path-tracing mode to Indiana Jones and the Great Circle™...
13 MIN READ
Four tiles of city scenes.
May 14, 2025

NVIDIA TensorRT Unlocks FP4 Image Generation  for NVIDIA Blackwell GeForce RTX 50 Series GPUs

The launch of the NVIDIA Blackwell platform ushered in a new era of improvements in generative AI technology. At its forefront is the newly launched GeForce RTX...
11 MIN READ
Apr 24, 2025

Fast Ray Tracing of Dynamic Scenes Using NVIDIA OptiX 9 and NVIDIA RTX Mega Geometry

Real-time ray tracing is a powerful rendering technique that can create incredibly realistic images. NVIDIA OptiX and RTX technology make this possible, even...
9 MIN READ

Conversational AI

See all
Jun 04, 2025

NVIDIA Speech AI Models Deliver Industry-Leading Accuracy and Performance

NVIDIA is driving state-of-the-art performance, efficiency, and accessibility in both speech AI and language models, setting the stage for innovations that are...
5 MIN READ
Jun 02, 2025

Scaling to Millions of Tokens with Efficient Long-Context LLM Training

The evolution of large language models (LLMs) has been marked by significant advancements in their ability to process and generate text. Among these...
7 MIN READ
data flywheel diagram
May 27, 2025

Upcoming Webinar: Supercharge Agentic AI with Scalable Data Flywheels

Join our live webinar on June 18 to see how NVIDIA NeMo microservices speed AI agent development.
1 MIN READ
image of jensen avatar
May 23, 2025

An Easy Introduction to LLM Reasoning, AI Agents, and Test Time Scaling

Agents have been the primary drivers of applying large language models (LLMs) to solve complex problems. Since AutoGPT in 2023, various techniques have been...
10 MIN READ
Students sitting around a computer.
May 07, 2025

Concept‑Driven AI Teaching Assistant Guides Students to Deeper Insights

In today's educational landscape, generative AI tools have become both a blessing and a challenge. While these tools offer unprecedented access to information,...
8 MIN READ
Apr 29, 2025

Spotlight: Personal AI Brings AI Receptionists to Small Business Owners with NVIDIA Riva

It's 10 p.m. on a Tuesday when the phone rings at the Sapochnick Law Firm, a specialized law practice in San Diego, California. The caller, a client of the...
6 MIN READ
Apr 22, 2025

NVIDIA GTC Training Labs Now Available On Demand

Missed GTC? This year’s training labs are now available on demand to watch anywhere, anytime.
1 MIN READ
Apr 18, 2025

Upcoming Event: NVIDIA Agent Toolkit Hackathon

Build a high-performance agentic AI system using the open-source NVIDIA Agent Intelligence toolkit — contest runs May 12 to May 23.
1 MIN READ

Edge Computing

See all
Jun 09, 2025

A Fine-tuning–Free Approach for Rapidly Recovering LLM Compression Errors with EoRA

Model compression techniques have been extensively explored to reduce the computational resource demands of serving large language models (LLMs) or other...
9 MIN READ
May 30, 2025

AI Brings Coral Reefs Into Focus

Researchers have unveiled a new AI model that can transform hard-to-see underwater images into clear, highly accurate 3D scenes. It can help ecologists more...
4 MIN READ
May 30, 2025

Telcos Across Five Continents Are Building NVIDIA-Powered Sovereign AI Infrastructure

AI is becoming the cornerstone of innovation across industries, driving new levels of creativity and productivity and fundamentally reshaping how we live and...
12 MIN READ
Decorative image.
May 19, 2025

NVIDIA TensorRT for RTX Introduces an Optimized Inference AI Library on Windows 11

AI experiences are rapidly expanding on Windows in creativity, gaming, and productivity apps. There are various frameworks available to accelerate AI inference...
9 MIN READ
ARC-Compact illustration.
May 18, 2025

Deploy AI-RAN at Cell Sites with NVIDIA ARC-Compact 

Wireless networks are the backbone of modern connectivity, serving billions of 5G users through millions of cell sites globally. The opportunities and benefits...
11 MIN READ
Decorative image.
Apr 16, 2025

Efficient Federated Learning in the Era of LLMs with Message Quantization and Streaming

Federated learning (FL) has emerged as a promising approach for training machine learning models across distributed data sources while preserving data privacy....
8 MIN READ
Apr 15, 2025

Event: Data Filtering Challenge for Training Edge Language Models

You’re invited to join the challenge. Develop and apply innovative data filtering techniques to curate datasets that enhance edge LM performance.
1 MIN READ
Decorative image.
Apr 11, 2025

Effortless Federated Learning on Mobile with NVIDIA FLARE and Meta ExecuTorch

NVIDIA and the PyTorch team at Meta announced a groundbreaking collaboration that brings federated learning (FL) capabilities to mobile devices through the...
12 MIN READ

Data Center / Cloud

See all
Jun 25, 2025

Powering the Next Frontier of Networking for AI Platforms with NVIDIA DOCA 3.0

The NVIDIA DOCA framework has evolved to become a vital component of next-generation AI infrastructure. From its initial release to the highly anticipated...
12 MIN READ
Jun 18, 2025

Improved Performance and Monitoring Capabilities with NVIDIA Collective Communications Library 2.26

The NVIDIA Collective Communications Library (NCCL) implements multi-GPU and multinode communication primitives optimized for NVIDIA GPUs and networking. NCCL...
11 MIN READ
Decorative image.
Jun 18, 2025

Benchmarking LLM Inference Costs for Smarter Scaling and Deployment

This is the third post in the large language model latency-throughput benchmarking series, which aims to instruct developers on how to determine the cost of LLM...
10 MIN READ
AI Virtual Camera video input and output.
Jun 17, 2025

Power Real-Time AI Media Effects with New AI Reference Apps on NVIDIA Holoscan for Media

Live media workflows are increasingly using AI microservices to augment production capabilities. However, advanced AI models are mostly hosted in the cloud,...
4 MIN READ
Decorative image.
Jun 12, 2025

Driving Toward Billion-Cell Analysis and Biological Breakthroughs with RAPIDS-singlecell

The future of cell biology and virtual cell models is dependent on measuring and analyzing data at scale. Single-cell experiments have been growing at an...
7 MIN READ
Jun 11, 2025

Introducing NVIDIA DGX Cloud Lepton: A Unified AI Platform Built for Developers

The age of AI-native applications has arrived. Developers are building advanced agentic and physical AI systems—but scaling across geographies and GPU...
6 MIN READ
Jun 10, 2025

How Modern Supercomputers Powered by NVIDIA Are Pushing the Limits of Speed — and Science

Modern high-performance computing (HPC) is enabling more than just quick calculations — it’s powering AI systems that are unlocking scientific...
6 MIN READ
Jun 05, 2025

Analyzing Baseboard Management Controllers to Secure Data Center Infrastructure

Modern data centers depend on Baseboard Management Controllers (BMCs) for remote management. These embedded processors enable administrators to reconfigure...
9 MIN READ