How NVIDIA Uses PyTorch

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.

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…

Frontier model pre-training has converged on mixture of experts (MoE), which is fundamentally changing what limits large-scale AI training. As compute per token…

Developers building 3D, design, simulation, robotics, and industrial digital twin applications need ways to bring physical AI capabilities into the tools and…

Useful quantum computers will require fault tolerant logical operations. Researchers are actively exploring many different quantum error correction (QEC) codes…

There are many ways to optimize code for GPUs. In this post, you’ll learn how kernel fusion can improve memory bandwidth and reduce kernel launch overhead…

Biomolecular structure prediction and co-folding with models like OpenFold3 are now mainstream, large-scale workloads powering drug discovery and protein design.

NVIDIA Omniverse NuRec is a neural reconstruction pipeline for building high-fidelity 3D representations of real-world environments from multisensor data such…

Generative AI workloads are rapidly outgrowing the memory and compute budget of single GPUs. For inference developers building media generation pipelines…

As AI systems move from single-turn interactions to coordinated multiagent workflows, low-latency inference becomes increasingly important.

Foundation models are reshaping computational biology. Pretrained on massive corpora of protein or genomic sequences, models such as ESM2 (a protein language…

As enterprise AI adoption scales, developers are increasingly forced to stitch together fragmented pipelines—separate models for text, vision…

This post is the third of a three-part series. See also Model Quantization: Concepts, Methods, and Why It Matters and Model Quantization: Post-Training…

AI agents are changing how you interact with your PC. Creators, developers, and AI enthusiasts are already using these agents extensively to assist with day-to…

Each wave of AI has created a new scaling law. Pretraining scaled intelligence through larger datasets, more parameters, and massively parallel GPU systems.

AI applications are moving beyond text generation to multimodal systems that can perceive, search, and reason across images, documents, video…

In production inference deployments, demand fluctuates over time, requiring inference replicas to scale elastically. However, cold-starting inference workloads…

Large language models (LLMs) are revolutionizing the financial trading landscape by enabling sophisticated analysis of vast amounts of unstructured data to…

NVIDIA CUDA 13.3 brings new capabilities and performance optimizations to developers across the CUDA ecosystem. The launch of NVIDIA CUDA Tile programming in…

The path from a trained AI model to production should be smooth, but rarely is. Many teams invest weeks fine-tuning models, only to discover that exporting to a…

This post is the second of a three-part series. See also Model Quantization: Concepts, Methods, and Why It Matters and Model Quantization: Turn FP8 Checkpoints…

For decades, computational biology has operated under a reductionist compromise. To fit complex biological systems into the limited memory of a single GPU…

Federated learning (FL) is no longer a research curiosity—it’s a practical response to a hard constraint: the most valuable data is often the least movable.

In March 2026, three LLM agents generated over 600,000 lines of code, ran 850 experiments, and helped secure a first-place finish in a Kaggle playground…

In a previous post, we introduced the Universal Sparse Tensor (UST), enabling developers to decouple a tensor’s sparsity from its memory layout for greater…

Higher-order optimization algorithms such as Shampoo have been effectively applied in neural network training for at least a decade. These methods have achieved…

The development of socially acceptable nuclear reactors requires that they are safe, clean, efficient, economical, and sustainable.

For decades, computational chemistry has faced a tug-of-war between accuracy and speed. Ab initio methods like density functional theory (DFT) provide high…

NVIDIA Ising is the world’s first family of open AI models for building quantum processors, launching with two model domains: Ising Calibration and Ising…

Training LLMs requires periodic checkpoints. These full snapshots of model weights, optimizer states, and gradients are saved to storage so training can resume…

Physical AI—AI systems that perceive, reason, and act in physically grounded simulated environments—is changing how teams design and validate robots and…

The Gemmaverse expands with the launch of the latest Gemma 4 multimodal and multilingual models, designed to scale across the full spectrum of deployments…

NVIDIA Groq 3 LPX is a new rack-scale inference accelerator for the NVIDIA Vera Rubin platform, designed for the low-latency and large-context demands of…

Computer-aided engineering (CAE) is shifting from human-driven workflows toward AI-driven ones, including physics foundation models that generalize across…

CUDA 13.2 arrives with a major update: NVIDIA CUDA Tile is now supported on devices of compute capability 8.X architectures (NVIDIA Ampere and NVIDIA Ada)…

Alibaba has introduced the new open source Qwen3.5 series built for native multimodal agents. The first model in this series is a ~400B parameter native vision…

The article discusses the use of NVFP4 low-precision model training to achieve higher throughput without sacrificing accuracy in AI model training.

The article discusses how the NVIDIA cuda. compute library enables Python developers to write high-performance GPU code without needing to resort to C++.

The article discusses how NVIDIA's hardware-software co-design significantly enhanced the inference performance of Sarvam AI's Sovereign 30B model, achieving a 4x speedup on NVIDIA Blackwell archit...

The article discusses NVIDIA TensorRT LLM AutoDeploy, a beta feature that automates the inference optimization process for large language models (LLMs).

Kimi K2. 5 is an advanced multimodal vision language model (VLM) developed by Kimi, optimized for various AI tasks.

The article discusses the challenges of Expert Parallel communication in training Mixture-of-Experts (MoE) models and introduces Hybrid-EP, an efficient communication solution that leverages NVIDIA...

The article discusses the Universal Sparse Tensor (UST), a framework designed to efficiently handle sparse tensors across various applications, including scientific computing and deep learning.

The article discusses the transition from the traditional two-phase API of the CUB library to a new single-call API introduced in CUDA 13. 1.

This article provides a detailed guide on implementing high-performance matrix multiplication using NVIDIA's cuTile framework in CUDA.

The article discusses NVIDIA's advancements in AI model inference performance through the Blackwell architecture, emphasizing improvements in token throughput per watt and the enhancements made to ...

The article discusses how recent upgrades to open source AI tools enhance the performance of small language models (SLMs) and diffusion models on NVIDIA RTX PCs.

The article discusses the latest software and model optimizations for NVIDIA DGX Spark, highlighting significant performance improvements in AI workflows.

The article discusses the NVIDIA Rubin platform, which introduces six new chips designed to create a powerful AI supercomputer.

NVIDIA introduces the Jetson T4000, enhancing AI and real-time reasoning for robotics and edge AI applications with up to 1200 FP4 TFLOPs of AI compute and 64 GB of memory.