Normal view

Received yesterday — 31 January 2026

Adaptive Inference in NVIDIA TensorRT for RTX Enables Automatic Optimization

26 January 2026 at 21:00
Deploying AI applications across diverse consumer hardware has traditionally forced a trade-off. You can optimize for specific GPU configurations and achieve...

Deploying AI applications across diverse consumer hardware has traditionally forced a trade-off. You can optimize for specific GPU configurations and achieve peak performance at the cost of portability. Alternatively, you can build generic, portable engines and leave performance on the table. Bridging this gap often requires manual tuning, multiple build targets, or accepting compromises.

Source

Received before yesterday

Building Generalist Humanoid Capabilities with NVIDIA Isaac GR00T N1.6 Using a Sim-to-Real Workflow 

8 January 2026 at 17:38
A robot thrTo make humanoid robots useful, they need cognition and loco-manipulation that span perception, planning, and whole-body control in dynamic environments. ...A robot thr

To make humanoid robots useful, they need cognition and loco-manipulation that span perception, planning, and whole-body control in dynamic environments. Building these generalist robots requires a workflow that unifies simulation, control, and learning for robots to acquire complex skills before transferring into the real world. In this post, we present NVIDIA Isaac GR00T N1.6…

Source

Accelerating LLM and VLM Inference for Automotive and Robotics with NVIDIA TensorRT Edge-LLM

8 January 2026 at 17:28
Large language models (LLMs) and multimodal reasoning systems are rapidly expanding beyond the data center. Automotive and robotics developers increasingly want...

Large language models (LLMs) and multimodal reasoning systems are rapidly expanding beyond the data center. Automotive and robotics developers increasingly want to run conversational AI agents, multimodal perception, and high-level planning directly on the vehicle or robot – where latency, reliability, and the ability to operate offline matter most. While many existing LLM and vision language…

Source

New Software and Model Optimizations Supercharge NVIDIA DGX Spark

5 January 2026 at 22:50
Since its release, NVIDIA has continued to push performance of the Grace Blackwell-powered DGX Spark through continuous software optimization and close...

Since its release, NVIDIA has continued to push performance of the Grace Blackwell-powered DGX Spark through continuous software optimization and close collaboration with software partners and the open-source community. These efforts are delivering meaningful gains across inference, training and creative workflows. At CES 2026, the latest DGX Spark software release, combined with new model…

Source

Accelerate AI Inference for Edge and Robotics with NVIDIA Jetson T4000 and NVIDIA JetPack 7.1

5 January 2026 at 22:10
NVIDIA Jetson T4000.NVIDIA is introducing the NVIDIA Jetson T4000, bringing high-performance AI and real-time reasoning to a wider range of robotics and edge AI applications....NVIDIA Jetson T4000.

NVIDIA is introducing the NVIDIA Jetson T4000, bringing high-performance AI and real-time reasoning to a wider range of robotics and edge AI applications. Optimized for tighter power and thermal envelopes, T4000 delivers up to 1200 FP4 TFLOPs of AI compute and 64 GB of memory, providing an ideal balance of performance, efficiency, and scalability. With its energy-efficient design and production…

Source

AI Factories, Physical AI, and Advances in Models, Agents, and Infrastructure That Shaped 2025

31 December 2025 at 17:30
Four-image grid illustrating AI agents, robotics, data center infrastructure, and simulated environments.2025 was another milestone year for developers and researchers working with NVIDIA technologies. Progress in data center power and compute design, AI...Four-image grid illustrating AI agents, robotics, data center infrastructure, and simulated environments.

2025 was another milestone year for developers and researchers working with NVIDIA technologies. Progress in data center power and compute design, AI infrastructure, model optimization, open models, AI agents, and physical AI redefined how intelligent systems are trained, deployed, and moved into the real world. These posts highlight the innovations that resonated most with our readers.

Source

Getting Started with Edge AI on NVIDIA Jetson: LLMs, VLMs, and Foundation Models for Robotics

11 December 2025 at 16:00
Running advanced AI and computer vision workloads on small, power-efficient devices at the edge is a growing challenge. Robots, smart cameras, and autonomous...

Running advanced AI and computer vision workloads on small, power-efficient devices at the edge is a growing challenge. Robots, smart cameras, and autonomous machines need real-time intelligence to see, understand, and react without depending on the cloud. The NVIDIA Jetson platform meets this need with compact, GPU-accelerated modules and developer kits purpose-built for edge AI and robotics.

Source

NVIDIA-Accelerated Mistral 3 Open Models Deliver Efficiency, Accuracy at Any Scale 

2 December 2025 at 18:10
The new Mistral 3 open model family delivers industry-leading accuracy, efficiency, and customization capabilities for developers and enterprises. Optimized...

The new Mistral 3 open model family delivers industry-leading accuracy, efficiency, and customization capabilities for developers and enterprises. Optimized from NVIDIA GB200 NVL72 to edge platforms, Mistral 3 includes: All the models were trained on NVIDIA Hopper GPUs and are now available through Mistral AI on Hugging Face. Developers can choose from a variety of options for deploying…

Source

💾

Model Quantization: Concepts, Methods, and Why It Matters

24 November 2025 at 19:23
Decorative image.AI models are becoming increasingly complex, often exceeding the capabilities of available hardware. Quantization has emerged as a crucial technique to address...Decorative image.

AI models are becoming increasingly complex, often exceeding the capabilities of available hardware. Quantization has emerged as a crucial technique to address this challenge, enabling resource-intensive models to run on constrained hardware. The NVIDIA TensorRT and Model Optimizer tools simplify the quantization process, maintaining model accuracy while improving efficiency.

Source

R²D²: Perception-Guided Task & Motion Planning for Long-Horizon Manipulation

4 November 2025 at 17:00
Traditional task and motion planning (TAMP) systems for robot manipulation use cases operate on static models that often fail in new environments. Integrating...

Traditional task and motion planning (TAMP) systems for robot manipulation use cases operate on static models that often fail in new environments. Integrating perception with manipulation is a solution to this challenge, enabling robots to update plans mid-execution and adapt to dynamic scenarios. In this edition of the NVIDIA Robotics Research and Development Digest (R²D²)…

Source

Make Sense of Video Analytics by Integrating NVIDIA AI Blueprints

3 November 2025 at 21:48
Decorative image.Organizations are increasingly seeking ways to extract insights from video, audio, and other complex data sources. Retrieval-augmented generation (RAG) enables...Decorative image.

Organizations are increasingly seeking ways to extract insights from video, audio, and other complex data sources. Retrieval-augmented generation (RAG) enables generative AI systems to use proprietary enterprise data. However, incorporating video content into these workflows introduces new technical hurdles, such as efficient ingestion, indexing, and maintaining compliance across diverse sources.

Source

Duos Technologies Achieves Positive EBITDA with 112% Quarterly Revenue Growth

13 November 2025 at 18:00

Duos Technologies Group, Inc. (Nasdaq: DUOT), a leader in intelligent technologies and digital infrastructure, has announced financial results for the third quarter ended September 30, 2025, reporting record revenue and positive adjusted EBITDA. The company remains on track to meet full-year guidance of $28 to $30 million, driven by growth in edge computing and energy services.

For the third quarter, Duos recorded $6.88 million in revenue, including $6.59 million in services, of which $5.15 million was related to its Asset Management Agreement (AMA) with New APR Energy. Total nine-month revenue reached $17.6 million, the highest in company history and reflecting more than 20% sequential growth.

During the quarter, Duos raised over $50 million to fund growth in the data center market and retired all debt. The company also reported improved gross margins and a reduced net operating loss. Key operational milestones included subsidiary Duos Edge AI’s new partnership with FiberLight to expand high-speed connectivity across underserved U.S. markets and the deployment of its sixth Edge Data Center, with nine additional sites planned for Q4, including the first out-of-state location in Illinois.

“I am very pleased with the continuous improvement in Duos’ results this year,” said Chuck Ferry, CEO of Duos Technologies. “Having embarked on the strategic shift to becoming a data center provider for the rapidly growing market for edge computing, we are well positioned to capture market share for products and services related to this important industry. I also welcome our new senior management and latest addition to our Board to assist in realizing the potential in 2026 and beyond.”

Duos also appointed Doug Recker as President and Corporate Officer, as well as Brian J. James to its Board of Directors, strengthening leadership and strategic alignment. The company was awarded U.S. Patent No. 12,404,690 B1 for its Entryway for a Modular Data Center and joined the Nomad Futurist Foundation as an Inspiration Sponsor, reinforcing its commitment to education and workforce development across the digital infrastructure ecosystem.

With approximately $25.8 million in backlog and near-term awards, Duos is reaffirming its 2025 revenue outlook and continues to focus on profitability, operational stability, and expansion within the edge and energy sectors.

The company’s management team hosted its third quarter 2025 earnings call on November 12, 2025, providing additional insight into results and strategy. A replay of the call is available in the investor section of the Duos Technologies website.

To learn more about Duos Technologies Group, Inc., visit www.duostech.com.

The post Duos Technologies Achieves Positive EBITDA with 112% Quarterly Revenue Growth appeared first on Data Center POST.

Europe’s Digital Infrastructure Enters the Green Era: A Conversation with Nabeel Mahmood at Capacity Europe

13 November 2025 at 16:00

Interview: Jayne Mansfield, ZincFive, with Nabeel Mahmood, Mahmood

At this year’s Capacity Europe conference in London – the epicenter for conversations shaping the digital infrastructure landscape – one theme cut through every panel and hallway exchange: Europe’s data future must be both powerful and sustainable.

To unpack what that really means for investors, operators, and policymakers, we sat down with technology executive and Top 10 Global Influencer Nabeel Mahmood, who spoke at the event about the region’s evolving data-center ecosystem.

“Demand is exploding across the UK and Europe,” Mahmood told us. “AI, edge compute, high-density GPU workloads, and hyperscale cloud deployments are all converging – and they’re forcing a rethink of what infrastructure looks like.” 

The Shift from Scale to Strategy

Mahmood’s central message was that the market’s priorities are shifting from ‘how much’ capacity to ‘how and where’ it’s built. Across the region, sustainability and energy resilience are no longer nice-to-have checkboxes; they’re becoming the foundation of investment decisions.

“Infrastructure used to be a race for megawatts,” he explained. “Now it’s a race for smarter, greener, and more sustainable megawatts.”

That shift is already visible in the UK, where annual data-center investment is projected to soar from roughly £1.75 billion in 2024 to £10 billion by 2029. While London remains dominant, new projects are spreading beyond the M25 as developers chase available power and faster permitting timelines.

Mahmood pointed out that “the UK’s declaration of data centers as critical national infrastructure is a step in the right direction – it signals recognition that digital infrastructure underpins everything from jobs to national competitiveness.”

Europe’s Tightrope: Power, Land, and Policy

Across continental Europe, the picture is similar but more constrained. The so-called FLAP-D markets – Frankfurt, London, Amsterdam, Paris, and Dublin – are nearing record-low vacancy rates, with take-up expected to hit 855 MW in 2025, up 22 % year-on-year.

“Grid capacity and land availability have become the new bottlenecks,” Mahmood said. “Those constraints are pushing investors to look at secondary markets – Milan, Nordic hubs, even parts of Southern Europe – where renewable energy integration and policy agility are improving.”

That migration is reshaping the map of European data infrastructure, with sustainability as the common denominator. Operators are incorporating liquid cooling, renewable sourcing, and battery-microgrid systems into new designs to support increasingly power-hungry AI clusters.

Why Power Chemistry Now Matters

In that context, Mahmood emphasized the critical role of next-generation battery technology – particularly nickel-zinc (Ni-Zn) – as a cornerstone of the sustainable data-center model.

“Battery systems are no longer just backup,” he said. “They’re becoming part of the strategic infrastructure footprint.”

Ni-Zn chemistry, he explained, offers a combination of high power density, safety, and circularity that aligns with Europe’s sustainability mandates. Unlike lithium-ion or lead-acid systems, Ni-Zn avoids thermal-runaway risks, reduces cooling needs, and offers recyclability benefits that fit the EU’s evolving battery-regulation framework.

“For operators, it’s not just an ESG checkbox,” Mahmood added. “It’s about freeing up space, cutting long-term costs, and demonstrating a credible pathway to low-carbon operations.”

A New Definition of Digital Infrastructure

Perhaps Mahmood’s most resonant message at Capacity Europe was philosophical: the way the industry defines “infrastructure” itself must evolve.

“Data centers aren’t just cost centers or tech assets,” he said. “They’re critical national infrastructure – pillars of the modern economy that touch climate policy, energy strategy, and digital sovereignty.”

That redefinition brings a new level of accountability. It means that as Europe scales for AI, cloud, and edge computing, the choices around power, cooling, materials, and footprint will determine not just commercial success but environmental integrity.

The Takeaway

Mahmood closed our conversation with a clear challenge to the industry:

“The digital-infrastructure boom sweeping through Europe must be anchored in responsible, resilient, and sustainable design. Adopting technologies like Ni-Zn isn’t just a technical upgrade – it’s a strategic differentiator. Those who embrace that mindset now will lead the next wave of growth.”

At Capacity Europe, optimism for digital expansion was everywhere – but so was a recognition that the future will belong to those who innovate responsibly. Mahmood’s vision distilled that reality perfectly: the next frontier of infrastructure isn’t just bigger. It’s smarter, greener, and built for permanence.

The post Europe’s Digital Infrastructure Enters the Green Era: A Conversation with Nabeel Mahmood at Capacity Europe appeared first on Data Center POST.

Duos Edge AI Joins Nomad Futurist Foundation as Inspiration Sponsor

6 November 2025 at 17:00

Duos Edge AI, Inc., a provider of adaptive, modular, and scalable Edge Data Center solutions, has joined the Nomad Futurist Foundation as an Inspiration Sponsor. The partnership reflects a shared commitment to advancing education, awareness, and leadership development within the digital infrastructure and edge technology sectors.

“At Duos Edge AI, we believe that education and access to technology are the foundation for innovation,” said Doug Recker, President of Duos and Founder of Duos Edge AI. “Part of our mission has been to expand that access and partnering with the Nomad Futurist Foundation allows us to help inspire the next generation and equip them with the knowledge and tools to shape the future.”

The Nomad Futurist Foundation is a global 501(c)(3) non-profit organization dedicated to empowering individuals from all backgrounds to explore and thrive in the digital future. As an Inspiration Sponsor, Duos Edge AI will support initiatives that help nurture the next generation of technology leaders through programs centered on digital infrastructure, AI, and edge computing.

Through this collaboration, Duos Edge AI and the Nomad Futurist Foundation will co-develop educational programs, community outreach initiatives, and awareness campaigns designed to broaden access to opportunities across the global technology landscape.

“We are thrilled to welcome Duos Edge AI as an Inspiration Partner,” said Nabeel Mahmood, Co-Founder of the Nomad Futurist Foundation. “Their commitment to innovation and education aligns perfectly with our mission to empower future leaders in the digital infrastructure space. Together, we can help more people discover and thrive in the careers that will define our digital future.”

The partnership reinforces both organizations’ dedication to creating a more inclusive and forward-looking digital ecosystem by connecting innovation, education, and opportunity.

To learn more about Duos Edge AI, visit www.duosedge.ai.

To learn more about the Nomad Futurist Foundation, visit nomadfuturist.org.

The post Duos Edge AI Joins Nomad Futurist Foundation as Inspiration Sponsor appeared first on Data Center POST.

❌