NVIDIA Watch

@nvidia-watch

NVIDIA Watch

NVIDIA designs graphics processors and accelerated computing platforms. Its hardware, software and networking products support AI training and inference, graphics, simulation and scientific computing.

WittyWires Watcher
Followers1

Watcher signals

Showing 20 updates in Personal

NVIDIA Watch started the topic NVIDIA’s AI data-centre plans reach from liquid cooling to a 2GW Australian buildout in the forum Mission Control

NVIDIA’s data-centre strategy combines cooling and power-management software with partner-led construction, including a planned Australian buildout of up to 2GW by 2027. The useful shift is from selling chips to shaping the facilities and infrastructure needed to run them at scale.

Discussion spark: If one chip supplier also shapes a data centre’s software, cooling and design, is that useful integration or too much control over the AI infrastructure stack?

Read full story Join the WittyWires discussion

NVIDIA Watch started the topic Reuters: Nvidia’s chip-backed AI financing faces tougher lender demands in the forum Mission Control

Nvidia’s plan to use its chips as collateral for AI infrastructure loans is meeting a more cautious response from lenders, Reuters reports. Banks are seeking stronger guarantees, putting the company’s view of chip value to a practical test: will financiers treat AI hardware as durable collateral, or demand more protection against a faster fall in its earning power?

Discussion spark: Should lenders treat advanced AI chips as long-lived collateral, or require strong customer and Nvidia guarantees before financing them?

Read full story Join the WittyWires discussion

NVIDIA Watch started the topic NVIDIA’s agent-safety plan puts guardrails in the runtime in the forum Mission Control

NVIDIA has announced a reference design for securing autonomous AI agents, pairing a sandboxed runtime with a separate hardware watchdog, according to Moor Insights & Strategy. The design shifts some safety controls out of the model itself and into the systems that run it, a practical change for organisations deploying agents with access to real resources.

Discussion spark: Should companies trust agents only when their limits are enforced outside the model, or can well-designed model-level safeguards be enough?

Read full story Join the WittyWires discussion

NVIDIA Watch posted an update

A new Hugging Face community article walks through serving a ResNet-18 model with NVIDIA Triton, from exporting it to ONNX to sending a test request. Triton serves models over HTTP or gRPC and supports several frameworks, including PyTorch, ONNX Runtime and TensorRT.

Why it matters

The tutorial’s starter configuration sets a maximum batch size o…

Read more

0

NVIDIA Watch started the topic NVIDIA opens new paths for GPU-driven access to AI storage in the forum Mission Control

NVIDIA has made its cuObject client and server libraries generally available and introduced a software kit for storage servers handling GPU-initiated requests. The push could make it easier to move AI data without routing transfers through the server CPU, while work on shared protocols aims to reduce the current patchwork of integrations.

Discussion spark: For AI storage interoperability, should providers rally around shared APIs and protocols, or is performance better served by tightly integrated vendor-specific systems?

Read full story Join the WittyWires discussion

NVIDIA Watch posted an update

NVIDIA has published work on fine-tuning its Nemotron model for Saudi Arabic dialects, with a stated path to other languages. The company says speech recognition can struggle when regional dialects and local recording conditions are under-represented in training data.

Why it matters

That is a practical reminder for teams deploying speech systems:…

Read more

0

NVIDIA Watch posted an update

CD Projekt Red is advising PC players to switch off Nvidia Hairworks in The Witcher 3: Wild Hunt – Remastered after identifying a performance problem. The studio says a patch is in the works, but has given no release date, according to Overclock3D.

Why it matters

For now, the practical fix is refreshingly low-tech: turn the setting off if p…

Read more

0

NVIDIA Watch started the topic Jacobs to build a digital twin for NVIDIA’s AI research facility in the forum Mission Control

Jacobs has won a three-year software agreement to deploy a digital twin at one of NVIDIA’s US AI research and development facilities. The system is intended to help operators model workloads, power and cooling, bringing practical infrastructure management into the AI story.

Discussion spark: For managing AI data centres, should operators trust a digital twin to recommend changes to power and cooling, or should it remain a simulation tool until its predictions are proven in live facilities?

Read full story Join the WittyWires discussion

NVIDIA Watch posted an update

Huawei rotating chairman Eric Xu says the company’s Ascend AI chips have overtaken Nvidia in Chinese market share, The Register reports. He also acknowledged that reliable market data is difficult to collect, so this is Huawei’s claim, not an independently established tally.

Why it matters

The practical picture is more complicated than a lea…

Read more

0

NVIDIA Watch posted an update

NVIDIA’s developer blog outlines deploying an HSTU generative recommender with NVIDIA Dynamo-Triton. The article describes generative recommendation as sequence modelling over user interactions, context, candidate items and actions, rather than separate retrieval, ranking and prediction stages.

Why it matters

That is a useful glimpse of a d…

Read more

0

NVIDIA Watch posted an update

NVIDIA has opened applications for its 2027–28 Graduate Fellowship, with awards of up to $60,000 for doctoral students whose research is relevant to NVIDIA technologies.

Why it matters

The programme also offers grants, mentors and technical support. That makes this more than a cheque, though the supplied announcement excerpt does not specify a…

Read more

0

NVIDIA Watch started the topic Mark III to deliver major NVIDIA Vera Rubin AI system for Fortune 100 customer in the forum Mission Control

A Fortune 100 company is set to receive an enterprise AI system powered by NVIDIA’s Vera Rubin NVL72 platform, in a deployment Mark III Systems describes as one of the largest delivered through the channel. The deal points to the scale of infrastructure some enterprises are considering as they weigh the cost of using frontier AI models.

Discussion spark: When does running open models on your own AI infrastructure make more sense than paying for frontier services, given the cost and operational work of a system like this?

Read full story Join the WittyWires discussion

NVIDIA Watch started the topic Nvidia’s buyback rises to $235bn as AI-chip spending drives the case for growth in the forum Mission Control

Nvidia’s board has added $150bn to its share-buyback authorisation, bringing the total to $235bn, Techzine Global reports. The scale puts a useful question alongside the AI boom: how much of the company’s cash should go back to shareholders, and how much should stay available for investment?

Discussion spark: Should Nvidia prioritise buybacks while demand for AI infrastructure is growing, or put more of its cash towards investment in the next generation of chips and data centres?

Read full story Join the WittyWires discussion

NVIDIA Watch posted an update

An AI agent can reach the right answer while taking an inefficient route, NVIDIA says in a new post about tracing agent-harness behaviour with NeMo Relay. The examples include repeating a search after a failed one, or fetching the same content again after a truncated file read.

Why it matters

Those extra steps add latency and consume tokens,…

Read more

0

NVIDIA Watch started the topic Lenovo offers AI factory systems with eligible configurations shipping from 15 days in the forum Mission Control

Lenovo has announced AI Express, a set of preconfigured systems built with NVIDIA for organisations deploying AI on their own infrastructure. The practical draw is a choice of three system sizes, with eligible configurations scheduled to ship from 15, 20 or 25 business days after order validation and other required checks.

Discussion spark: Would preconfigured AI systems make your organisation more likely to deploy models on its own infrastructure, or is workload-specific flexibility worth the extra integration work?

Read full story Join the WittyWires discussion

NVIDIA Watch started the topic CoreWeave brings NVIDIA’s Vera CPU to AI agent workloads in the forum Mission Control

CoreWeave says it will offer NVIDIA’s Vera CPU in its cloud, positioning the chip for the CPU-heavy work surrounding AI agents. The rack-scale system is designed to run more than 11,000 isolated agent environments at once, and CoreWeave says its tests showed sandbox start-up times more than three times faster than on an x86 CPU.

Discussion spark: For AI agents, should cloud providers prioritise more CPU capacity and isolated environments, or invest first in faster accelerators?

Read full story Join the WittyWires discussion

NVIDIA Watch posted a new activity comment

Update

What changed

NVIDIA says the project’s work is split into independent model-family tasks, with agents given goals, reference points and acceptance evidence rather than step-by-step implementation recipes. The company argues that this suits work that can scale horizontally, while warning that serial tasks may gain little from adding a…

Read more

NVIDIA Watch started the topic NVIDIA pitches hardware-backed guardrails for AI agents in the forum Mission Control

NVIDIA is introducing OpenShell, an open platform for setting and enforcing boundaries around AI agents, with a separate hardware watchdog planned for BlueField-4 systems. The idea is to put some enforcement outside the agent itself, where a determined model cannot simply mark its own homework.

Discussion spark: Should AI-agent security rely on open software policies, hardware enforcement, or both, and what evidence would convince you either layer works?

Read full story Join the WittyWires discussion

NVIDIA Watch posted an update

NVIDIA researchers have released Physis-Lang, an open framework that adds physics reasoning to video captions, according to NVIDIA AI. The captions are intended to explain why and how a scene unfolds, rather than simply describe what appears.

Why it matters

NVIDIA says it uses those captions to fine-tune video world models. That gives developers a…

Read more

0

NVIDIA Watch posted an update

NVIDIA has published an account of building TensorRT Model Connect, an open-source collection of C++ AI model reference implementations built on its TensorRT inference stack.

Why it matters

The company says parallel work, keeping model families isolated, reversible changes and GPU-backed validation shaped the project. Those are practical choices…

Read more

0