❌

Normal view

ModelExpress: Distributing Model Artifacts at the Speed of Light

24 July 2026 at 16:45
Every byte moved has a cost. As model checkpoints grow to hundreds of gigabytes or even a terabyte, that cost adds up quickly. To make things even worse, moving...

Every byte moved has a cost. As model checkpoints grow to hundreds of gigabytes or even a terabyte, that cost adds up quickly. To make things even worse, moving these model weights around the cluster is extremely common. For instance, a cold start may pull weights from remote storage into GPU memory; autoscaling and rolling updates must populate each new replica; and RL post-training continuously…

Source

Inside NVIDIA Rubin GPU Architecture: Powering the Era of Agentic AI

23 July 2026 at 18:15
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....

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. These factories are now tasked with powering agentic workflows that reason, plan, use tools, verify intermediate results, and execute complex multistep tasks across vast contexts. Agentic workloads are not defined by a single prompt…

Source

Setting a World Record for MoE Pre-Training on NVIDIA GB300 NVL72

23 July 2026 at 18:30
Decorative image.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...Decorative image.

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 falls, communication increasingly determines how efficiently models scale across thousands of GPUs. NVIDIA GB300 NVL72 set a world record for pre-training DeepSeek-V3 671B at 1,648 TFLOPs per GPU, showing how advances across the entire AI…

Source

NVIDIA Vera CPU: Olympus Cores Built for Maximum Single-Thread Performance in Agentic AI

23 July 2026 at 18:00
Vera CPU image.Agentic AI shifts more of the critical execution path onto the CPU. Agents operate in sandboxes to execute code, invoke tools, retrieve context, interact with...Vera CPU image.

Agentic AI shifts more of the critical execution path onto the CPU. Agents operate in sandboxes to execute code, invoke tools, retrieve context, interact with databases, and analyze results before returning information to the model. As these loops run concurrently across an AI factory, CPU performance increasingly shapes both per-agent responsiveness and overall factory throughput.

Source

Scaling Agentic AI Factories Through Extreme Co-Design with NVIDIA BlueField

16 July 2026 at 16:00
Decorative image.Agentic AI changes the infrastructure pattern for AI factories. One request can trigger many model calls, tool calls, memory lookups, policy checks, storage...Decorative image.

Agentic AI changes the infrastructure pattern for AI factories. One request can trigger many model calls, tool calls, memory lookups, policy checks, storage accesses, and network transfers before a final answer is produced. As more agents run at once and carry context across steps, users, tools, services, and sessions, infrastructure must move, protect, retrieve, and reuse data fast enough to keep…

Source

NVIDIA Ising Decoding Cuts Color Code Logical Error Rates by Over 300x

13 July 2026 at 19:00
Useful quantum computers will require fault tolerant logical operations. Researchers are actively exploring many different quantum error correction (QEC) codes...

Useful quantum computers will require fault tolerant logical operations. Researchers are actively exploring many different quantum error correction (QEC) codes to enable this, improving the Logical Error Rates (LER) of Quantum Processing Units (QPUs). While it is well understood how to run logical operations with surface codes (which belong to the topological code family) via lattice surgery…

Source

AI Model Co-Design: Hardware-Friendly LLM Design

10 July 2026 at 16:36
AI performance comes down to three dimensions:Β  Accuracy: How well the model reasons and produces outputs Throughput: How many tokens per second a...

AI performance comes down to three dimensions: Deployments must balance all three: High accuracy is wasted if responses are slow, and raw throughput means little if each user’s experience is laggy. Practical systems therefore optimize accuracy, throughput, and interactivity together. This post focuses on throughput and interactivity, and how model-design choices shape both without…

Source

Enhancing Goodput in Large-Scale LLM Training with Nonuniform Tensor Parallelism

6 July 2026 at 21:44
Deocrative image.Training LLMs at massive scale brings unique infrastructure challenges, especially as jobs span thousands of GPUs and run for extended periods. The longer these...Deocrative image.

Training LLMs at massive scale brings unique infrastructure challenges, especially as jobs span thousands of GPUs and run for extended periods. The longer these jobs run, the greater the likelihood of encountering unscheduled interruptions or resource fluctuations. Even infrequent device unavailability can have outsized effects on tightly interconnected clusters, resulting in slowdowns for a given…

Source

Hardware-Rooted AI Security That Won’t Slow You Down

2 July 2026 at 21:25
Decorative image.AI has transformed how organizations operate, driving unprecedented levels of productivity and innovation. However, AI adoption can be impeded by concerns...Decorative image.

AI has transformed how organizations operate, driving unprecedented levels of productivity and innovation. However, AI adoption can be impeded by concerns surrounding data privacy, sovereignty and how to secure data while it is in use, or during inference and engagement with AI models. NVIDIA Confidential Computing (CC) was engineered to be a secure and performant solution for the era of agentic…

Source

Designing GPU-Accelerated Query Engines with NVIDIA GQE

30 June 2026 at 17:36
Decorative image.GPU-accelerated query engines are often constrained by memory and I/O bandwidth. NVIDIA hardware advancesβ€”including high bandwidth memory (HBM), NVIDIA...Decorative image.

GPU-accelerated query engines are often constrained by memory and I/O bandwidth. NVIDIA hardware advancesβ€”including high bandwidth memory (HBM), NVIDIA NVLink-C2C, and dedicated decompression engines featured in NVIDIA GB200 NVL4β€”help remove these bottlenecks by increasing effective storage capacity, accelerating data movement between CPUs and GPUs, and speeding data access without consuming…

Source

How to Govern Autonomous Agents in Enterprise AI FactoriesΒ 

29 June 2026 at 15:50
Decorative image.AI agents are quickly moving beyond chat. They inspect code, run tests, read documents, search knowledge bases, query internal systems, and operate for hours on...Decorative image.

AI agents are quickly moving beyond chat. They inspect code, run tests, read documents, search knowledge bases, query internal systems, and operate for hours on behalf of a user. This unlocks productivity, but can also give agents access to sensitive enterprise data and the ability to complete tasks and take action across business systems, making a secure, governed environment essential.

Source

Deploy a Production-Ready NVIDIA AI-Q Blueprint on Oracle Cloud Infrastructure

26 June 2026 at 19:00
Decorative image.AI agents have changed a lot in the last two years. The first could only answer one question at a time. Then came multi-turn chat, where the model could keep...Decorative image.

AI agents have changed a lot in the last two years. The first could only answer one question at a time. Then came multi-turn chat, where the model could keep some context across a session. Today, we have long-horizon agents. Systems that plan many steps, split work between sub-agents, keep context across a long task, and run tools in a safe sandbox. The NVIDIA AI-Q Blueprint is an open source…

Source

❌