A naukri.com initiative
AR News
Digitaltrends
4d
320
Image Credit: Digitaltrends
Nvidia’s next-gen GPUs could get a major design change, here’s why that’s good news
- Nvidia is collaborating with TSMC to develop next-gen GPUs using chiplet technology.
- The shift to chiplet-based design offers improved performance, scalability, and cost efficiency.
- Nvidia's Rubin GPUs will leverage TSMC's N3P process node and advanced packaging technologies.
- The adoption of chiplet technology is expected to bring breakthrough advancements in GPU design.
Read Full Article
19 Likes
Insider
4d
32
Image Credit: Insider
Meet a decades-old software company hitching a ride on the Nvidia rocket ship
- DataDirect Networks (DDN), a decades-old computing company, is benefiting from the AI boom.
- DDN offers crucial data storage for AI data centers, specializing in fast access to stored data for GPU users.
- The company has been working with Nvidia for eight years and is now seeing exponential success due to the scalability of AI storage systems.
- DDN recently raised $300 million from Blackstone, valuing the company at $5 billion and positioning it for further expansion.
Read Full Article
1 Like
Insider
4d
20
Image Credit: Insider
CoreWeave is on the cusp of a huge IPO — but it could face an expensive hardware problem
- CoreWeave, a startup planning a major tech IPO, may face challenges due to rapid advancements rendering its Hopper-based Nvidia GPUs obsolete.
- The AI technology upgrades are impacting the useful life of CoreWeave's GPUs, particularly as Nvidia rolls out superior products like Blackwell.
- Amazon's recent adjustment of server useful life due to AI progress suggests a potential issue for other tech firms, including CoreWeave.
- CoreWeave, a neocloud specializing in generative AI workloads, has deployed over 250,000 Nvidia GPUs, but faces the threat of rapid obsolescence.
- The rapid pace of AI technology advancement can quickly make existing gear less valuable, as seen with the transition from Hopper to Blackwell GPUs.
- Companies like CoreWeave may have to adjust financials to account for the fast-evolving product landscape and depreciate assets accordingly.
- Amazon decreased server useful life from six to five years, potentially impacting operating income, with similar moves expected from other tech firms.
- CoreWeave's IPO filing acknowledges the need for continuous platform upgrades and uncertainty in estimates related to infrastructure components' useful life.
- While CoreWeave benefited from an increased estimated useful life of computing gear in 2023, recent developments may lead to earnings challenges.
- CoreWeave's response to potential adjustments in the useful life of its equipment remains unclear as it navigates the complexities of planning a high-stakes IPO.
Read Full Article
1 Like
TronWeekly
4d
284
Image Credit: TronWeekly
Hedera’s HBAR Rockets 144%? NVIDIA & Intel’s AI Breakthrough Revealed!
- HBAR targets $0.200, with potential for a 144% rally if it breaks key resistance.
- NVIDIA and Intel are building an AI governance solution on Hedera, using its consensus service for immutable AI computation ledgers.
- AUDD Stablecoin and RWA_xyz plan integrations with HBAR, expanding its use in the multi-trillion-dollar stablecoin and RWA tokenization markets.
- Hedera's HBAR price target remains at around $0.504 level, potentially soaring to over a 144% rally if momentum continues.
Read Full Article
17 Likes
Medium
4d
341
Image Credit: Medium
NVIDIA’s AI Power Move; You May Not Know This, But You Should!
- NVIDIA is aggressively pursuing AI robotics and announced Gr00t N1, a foundational model for robots.
- NVIDIA's Omniverse/Isaac Gym provides a digital playground for training AI robots in simulations, reducing costs and accelerating learning.
- NVIDIA is entering the high-performance consumer-end AI computing market with two new products to compete with Apple's M4 and M3 Ultra series.
- NVIDIA is focusing on AI data centers, using modular computing units and introducing the Vera Rubin platform for increased compute power and memory efficiency.
- NVIDIA's announcements indicate that AI will continue to integrate into various industries, and their influence in AI hardware is undeniable.
Read Full Article
20 Likes
Siliconangle
5d
232
Image Credit: Siliconangle
Cerebras Systems faces delays in IPO as CFIUS review remains unresolved
- The initial public offering (IPO) of Cerebras Systems Inc., an artificial intelligence chip-making company, is facing delays due to unresolved national security clearance and the review by the Committee on Foreign Investment in the United States (CFIUS).
- Cerebras executives are waiting for key appointments to be made by the White House and for the CFIUS review to be completed, as approval is needed for the IPO due to an investment by G42, which has ties to China.
- Cerebras produces specialized computer chips for AI and high-performance computing workloads, claiming they are more powerful and cost-effective than Nvidia Corp.'s graphics processing units.
- The delay in the IPO is also attributed to concerns that CFIUS staffers might be hesitant to approve large deals during a time of government employee cuts, and due to previous concerns about G42's China links.
Read Full Article
13 Likes
Hackaday
5d
248
Image Credit: Hackaday
Brazilian Modders Upgrade NVidia Geforce GTX 970 to 8 GB of VRAM
- A group of Brazilian modders have found a way to upgrade the NVidia GeForce GTX 970 to 8 GB of VRAM.
- The original 512 MB, 7 Gbps GDDR5 memory modules are replaced with 1 GB, 8 Gbps chips and a resistor is added to make the GPU recognize the higher density VRAM ICs.
- This upgrade does not fix the fundamental split VRAM issue of the ASIC, but it does provide access to 7 GB of faster, higher-density VRAM.
- The benchmark performance of the upgraded GTX 970 shows a significant increase, with nearly a doubling in the Unigine Superposition score.
Read Full Article
14 Likes
VentureBeat
5d
187
Image Credit: VentureBeat
Beyond transformers: Nvidia’s MambaVision aims to unlock faster, cheaper enterprise computer vision
- Nvidia's MambaVision is a Mamba-based model family for computer vision and image recognition tasks, aiming to improve efficiency and accuracy of vision operations at lower costs.
- MambaVision combines the efficiency of Mamba with the modeling power of Transformers, employing both attention mechanisms and convolutional approaches to process visual information.
- The newly released MambaVision models on Hugging Face have been trained on larger datasets, offering better performance and capabilities for diverse and complex tasks.
- MambaVision's benefits include reduced inference costs, potential for edge deployment, improved downstream task performance, and simplified deployment with Hugging Face integration.
Read Full Article
11 Likes
Arstechnica
5d
302
Image Credit: Arstechnica
No cloud needed: Nvidia creates gaming-centric AI chatbot that runs on your GPU
- Nvidia has released an experimental G-Assist AI, a gaming-centric chatbot.
- G-Assist runs locally on Nvidia GPUs to optimize PC and maximize gaming performance.
- It provides system stats, tweaks settings, creates data charts, and can overclock GPUs.
- The chatbot is available in the Nvidia desktop app and is still in the experimental stage.
Read Full Article
18 Likes
The Robot Report
5d
114
Teradyne Robotics, NVIDIA show latest AI Accelerator toolkit at GTC
- Teradyne Robotics, in collaboration with NVIDIA, unveiled the AI Accelerator toolkit at NVIDIA GTC 2025.
- The AI Accelerator is designed to enable the development of AI-powered applications, bringing AI acceleration to Universal Robots' (UR) PolyScope X software.
- The toolkit equips robots with the ability to learn, adapt, and make informed decisions based on sensory input, improving their problem-solving capabilities in complex use cases.
- The AI Accelerator also provides developers an extensible platform to build and de-risk applications, reducing time to market.
Read Full Article
6 Likes
Siliconangle
5d
28
Image Credit: Siliconangle
Turning hardware into systems: How Dell and Nvidia are creating a new IT paradigm with the AI Factory
- Dell and Nvidia are collaborating to create the AI Factory, offering end-to-end AI enterprise solutions for training and running models.
- The AI Factory signifies a shift towards systems driving productivity and AI innovation rather than just hardware solutions.
- During the SC24 conference in November, Dell introduced enhancements to AI Factory in collaboration with Nvidia, focusing on AI performance tools.
- Dell added Nvidia GPUs to the PowerEdge XE9685L, enhancing GPU density for AI and HPC workloads.
- They unveiled Agentic RAG with Nvidia, a process to scale AI for multiple use cases, improving performance and enabling digital assistants.
- RAG, a technique discussed at Nvidia's GTC gathering, is a key trend for generative AI scale-out.
- Dell integrated Nvidia Grace CPUs and Blackwell GPUs in PowerEdge systems for AI workloads.
- Nvidia's focus on system balance and high memory bandwidth is crucial for AI computing infrastructure.
- Infrastructure for the AI Factory will involve retooling data center technologies, and Dell is working on minimizing network complexity for enterprise customers.
- Dell and Nvidia offer solutions to help companies quickly deploy tested and optimized AI infrastructure at scale.
Read Full Article
1 Like
Macdailynews
5d
171
Image Credit: Macdailynews
Apple joins AI data center race, orders $1 billion in Nvidia systems
- Apple has recently entered the AI data center race by ordering approximately $1 billion worth of Nvidia systems for its data center in Reno, Nevada.
- The move signifies Apple's recognition of the importance of AI data centers, with a focus on generative AI applications.
- Apple is collaborating with server builders Dell Technologies and Super Micro Computer for its large server cluster to support AI applications.
- The strategy shift was prompted by delays in launching AI features, particularly the AI-enabled Siri digital assistant.
- Apple's executive team has undergone changes in response to challenges in releasing AI features, with criticism aimed at the leadership's lack of vision.
- The company's efforts to catch up in the AI space involve partnerships with companies like Google and OpenAI and focus on enhancing on-device AI capabilities.
- Critics suggest that Tim Cook's leadership lacks vision and innovation compared to Steve Jobs, prompting Apple to resort to a 'vaporware' strategy to buy time.
- Apple's marketing efforts, including unveiling new AI features at WWDC, are seen as attempts to position the company as an innovator in AI despite lagging behind competitors.
- The transition to catch up with rivals in the AI domain requires significant investments in data center infrastructure and technology development.
- While Apple is currently playing catch-up in AI, it is hoped that future innovations and catch-up efforts will help the company compete and potentially surpass its rivals in the AI space.
Read Full Article
10 Likes
Eu-Startups
5d
401
“How Europe Can Compete in Quantum & AI on the Global Stage” – Live panel at the EU-Startups Summit 2025!
- The EU-Startups Summit 2025 in Malta will bring together founders, startup enthusiasts, and investors for talks and networking.
- A panel at the summit will discuss Europe’s global position in AI and quantum computing, addressing funding challenges and regulatory roles.
- Panelists include Sabrina Maniscalco from Algorithmiq, Rodrigo Hernandez from Multiverse Computing, and Serge Lemonde from NVIDIA.
- Algorithmiq focuses on quantum computing for life sciences, Multiverse Computing specializes in quantum AI, and NVIDIA supports startups in AI and quantum.
- Malta Enterprise is the headline sponsor, supporting economic growth and innovation, while Scytale and Airwallex are among the sponsors offering compliance and financial solutions.
- EUROCROWD promotes crowdfunding for SMEs, the M. Demajo Group is a key business player in Malta, and the European Innovation Council drives disruptive science into commercial innovation.
Read Full Article
24 Likes
Guardian
5d
384
Image Credit: Guardian
From spit to bankruptcy: the rise and fall of 23andMe
- Genetic testing firm 23andMe filed for bankruptcy, with the CEO stepping down after failed buyout attempts and a decline from its peak valuation of $5.8bn in 2021.
- The company initiated voluntary Chapter 11 proceedings to facilitate a sale process to maximize business value, assuring customers of no change in data management amidst past data breach incidents.
- Customers are left wondering about the fate of the genetic data collected by 23andMe over the years, as the company's promise of personalized medicine based on genetic code remains unfulfilled.
- Meanwhile, Nvidia unveiled new powerful chips enhancing AI computing capacity, with developments hinting at a future fusion of AI and robotics, exemplified by a Disney-partnered robot showcased at the conference.
- As AI companies venture into agentic models and anticipation grows for artificial general intelligence, the tech landscape is gearing up for the convergence of advanced AI capabilities with robotics.
- On the entertainment front, AI is being utilized to create short-form fiction for platforms like Instagram, where creators blend human storytelling with AI-generated imagery to engage audiences in novel ways.
- Artists raise concerns over the use of AI in art production, citing intellectual property issues, as seen in the Atlantic's tool for authors to check pirated content and legal battles surrounding AI models trained on copyrighted material.
- Despite debates over AI's impact on creative industries, AI-enabled creators like @HolyFool36 on Instagram are leveraging technology to rapidly produce and share fictional content, opening new avenues for serialized storytelling.
- From 23andMe's bankruptcy to the intersection of AI and robotics, tech developments continue to shape various sectors, highlighting the evolving landscape of innovation, data privacy concerns, and the transformative power of artificial intelligence.
Read Full Article
23 Likes
Digitaltrends
5d
57
Image Credit: Digitaltrends
Nvidia G-Assist uses AI to configure game settings so you don’t have to
- Nvidia's G-Assist AI assistant is now available on the Nvidia app for GeForce RTX desktop users.
- The assistant aims to optimize PC settings by helping users switch between different settings depending on the game being played.
- G-Assist can be prompted through text or voice commands and does not require an internet connection.
- Nvidia has provided official plugins and a GitHub repository for users to create their own community plugins.
Read Full Article
3 Likes
For uninterrupted reading, download the app