A naukri.com initiative
AR News
Insider
4w
348
Image Credit: Insider
Jensen Huang says the 3 elements of AI scaling are all advancing. Nvidia's Blackwell demand will prove it.
- Jensen Huang, CEO of Nvidia, stated that the three elements of scaling in AI are continuing to advance.
- There have been concerns about AI progress slowdown and model scaling.
- Huang mentioned that the scaling of foundation model pre-training is intact and continuing.
- The demand for Nvidia's Blackwell chips is a result of the need for higher-powered compute.
Read Full Article
20 Likes
Analyticsindiamag
4w
94
Image Credit: Analyticsindiamag
AI Scaling Laws Crumble Under Token Pressure
- Microsoft’s Satya Nadella tackled the AI scaling challenge at Microsoft Ignite.
- The scaling laws in AI are not physical laws but empirical observations.
- Efforts in areas such as model architectures, data regimes and system architectures are needed to push innovation.
- Microsoft is working closely with NVIDIA on scaling test-time computing for OpenAI.
- 'Tokens per watt plus dollar' is Microsoft's new metric to reflect the efficiency of generating tokens.
- Data centres are now viewed as a product by NVIDIA rather than just GPUs.
- Electricity, supply chain and capital are major challenges when it comes to scaling computing.
- OpenAI plans to partner with TSMC and Broadcom to launch its first in-house AI chip by 2026.
- Major inference chip makers like Groq, Sambanova and Cerebras have improved performance.
- NVIDIA is working with its suppliers to cool overheating racks.
Read Full Article
5 Likes
Tech Story
4w
214
Nvidia Nearly Doubles Revenue in Q3 Amid Strong AI Demand and GPU Sales
- Nvidia Corporation reports an impressive Q3 FY 2025 revenue of $18.12 billion, almost doubling its sales from the previous year.
- The data center division accounts for over 76% of overall sales, driven by increased demand for Nvidia's GPUs, particularly the H100 processors used in AI models.
- Nvidia's gross margins reach 72.6%, highlighting its profitability amidst high demand.
- The company's global expansion and strategic investments in semiconductor technology position it as a major player in the AI industry.
Read Full Article
12 Likes
Skarredghost
4w
201
Image Credit: Skarredghost
Hands-on: Varjo XR-4 Focal Edition takes passthrough to the next level
- The Varjo XR-4 is a high-resolution headset for enterprises, primarily for companies where graphical resolution is of paramount importance.
- The device features 3840 x 3744 resolution per eye, and a 120° x 105° field of view.
- The Varjo XR-4 Focal Edition features an equivalent pixel density of 51 PPD.
- Autofocus actions from the eye-tracking of the headset drives the autofocus of the 20 MP passthrough cameras.
- This provides a passthrough with an incredibly crisp texture and a 51 PPD pixel density.
- Retina resolution is generally considered to be around 60 PPD, so the Varjo XR-4 Focal Edition is very close to what human vision requires.
- The high resolution, however, is only available inside a small, rectangular region, and there is a noticeable lag when moving the eyes from one object to another.
- Reducing the lag and blurring the edges of the high-definition focus area could help this headset provide clearer visuals.
- Passthrough vision could presumably become as crisp as our regular vision, enabling us to complete tasks without having to remove our headset entirely.
- However, the device costs €10,000, which could pose a potential hurdle for many enterprises.
Read Full Article
12 Likes
TechCrunch
4w
1.2k
Image Credit: TechCrunch
Nvidia’s CEO defends his moat as AI labs change how they improve their AI models
- Nvidia's CEO, Jensen Huang, defended the company's position amidst changing AI model improvement methods.
- Huang acknowledged the significance of 'test-time scaling' and how it could play a larger role in Nvidia's business.
- The increasing emphasis on AI inference poses a challenge for Nvidia as competitors develop faster AI inference chips.
- Huang assured investors that AI model developers are still improving their models through increased compute and data during pretraining.
Read Full Article
11 Likes
Pymnts
4w
272
Image Credit: Pymnts
After Posting 94% Revenue Jump, Nvidia CEO Says Company Has Room to Scale
- Nvidia CEO, Jensen Huang, believes that the future of the company, and AI in general, is promising.
- The company's session outcomes for Q3 exceeded market expectations, and we'll see Nvidia enter new markets in the next few years.
- Huang referred to Nvidia's data centres as 'AI factories,' that are designed to handle the computational demands of AI.
- Nvidia has built Blackwell architecture to meet the needs of the evolving AI landscape. Seven custom chips provide support for various CPU options.
- Despite concerns about the scaling of LLMs, particularly with reference to a potential demand-side slowdown, Huang believes that growth opportunities are still available.
- Huang emphasised the shift from traditional coding to machine learning as a fundamental shift that will support AI workloads.
- Huang cited the emergence of generative AI as a whole new market segment and opportunities in intelligence, digital intelligence, artist intelligence and legal intelligence.
- Other growth in the industrial AI sector was generated from the use of Nvidia's Omniverse platform to improve operating efficiency and automate workflows.
- Collaboration with major cloud service providers such as Dell, Corweave, Oracle, Microsoft, and Google enabled Nvidia to execute and integrate their systems successfully.
- Nvidia is expecting modernization to continue, estimates several more years of growth, and believes the best days are yet to come.
Read Full Article
16 Likes
Medium
4w
8
Image Credit: Medium
NVIDIA: The Powerhouse Behind AI, Gaming, and Data Centers
- NVIDIA initially produced graphics cards for video games, with the GeForce 256 launched in 1999 as the first marketed GPU.
- The company later expanded into AI and deep learning, thanks in part to GPUs being ideal for the sequential tasks required by machine learning.
- NVDA’s Tesla and A100 GPUs are some of the best for training large AI models and are used in natural language processing, image recognition, robotics, and more.
- NVIDIA’s acquisition of ARM Holdings is still awaiting regulatory approval, but would give the company greater control over the design of chips used in smartphones, tablets, and other devices.
- The company has established a foothold in cloud-based AI workloads, powering advancements in autonomous vehicles and next-gen entertainment platforms.
- Gaming remains a core part of NVIDIA’s business, and its GeForce series of GPUs dominate the market.
- The autonomous vehicle industry has also seen significant strides from the company, with NVIDIA DRIVE designed to power self-driving cars.
- NVIDIA’s data center revenue contributed significantly to its overall growth, and the company’s NVIDIA Q3 earnings 2024 were above expectations.
- Competition in the GPU market is intensifying, particularly with the rise of AMD and Intel, and regulatory scrutiny is growing over the company’s acquisition of ARM.
- NVIDIA has positioned itself well for the future, with investment in AI, cloud computing, and autonomous vehicles expected to play an increasingly vital role.
Read Full Article
Like
Pymnts
4w
250
Image Credit: Pymnts
Report: Elon Musk’s xAI Raises $5 Billion and Doubles Valuation
- Elon Musk’s AI startup xAI raised $5 billion in a funding round, valuing the company at $50 billion.
- This brings xAI's total funding for the year to $11 billion and more than doubles its valuation from $24 billion to $50 billion.
- The funds will be used to add 100,000 Nvidia chips for training AI models.
- Menlo Ventures predicts that enterprise spending on generative AI will increase from $2.3 billion in 2023 to $13.8 billion in 2024.
Read Full Article
15 Likes
Siliconangle
4w
152
Image Credit: Siliconangle
Nvidia posts another strong earnings beat as customers race to deploy next-gen Blackwell GPUs
- Nvidia delivered better-than-expected Q3 earnings with earnings before certain costs at 81 cents per share, beating the Wall Street’s projection of 75 cents.
- Revenue rose 94% YoY to $35.08bn, ahead of the $33.16 billion analyst target.
- Net income rose to $19.3 bn in Q3, up from just $9.24 bn in the year-ago quarter.
- In Q4, Nvidia said it’s expecting to see revenue of around $37.5bn at the midpoint of its guidance range, just above the analysts’ consensus estimate of $37.08bn, implying YoY growth of 70%.
- The growth stems from Nvidia’s data center business, attributed $30.8 billion of its sales, up 112%, while sales of GPUs for gaming delivered $3.28 billion.
- Most of the growth is driven by incredible demand from cloud infrastructure providers and other enterprises for its iconic graphics processing units, which power the vast majority of AI workloads today.
- Nvidia's Blackwell GPUs have been shipped to several major customers with shipments expected to ramp up significantly.
- Blackwell shipments are expected to exceed supply for several quarters in fiscal 2026.
- The company recently announced the launch of a new and improved version of the H200 GPUs called the H200 NVL.
- Third Bridge analyst Lucas Keh expects Nvidia to be able to maintain sequential growth for some time to come.
Read Full Article
9 Likes
Guardian
4w
429
Image Credit: Guardian
Nvidia earnings: AI chip leader shows no signs of stopping mammoth growth
- Nvidia reported quarterly revenue of $35.08bn, surpassing expectations of $33.15bn, with profits doubling year-over-year.
- The company projected a 70% increase in revenue for the coming quarter.
- Nvidia's CEO, Jensen Huang, highlighted the shift from coding to machine learning and the potential of generative AI.
- Demand for Nvidia's Blackwell GPU chips remains strong, positioning the company as a major player in the AI industry.
Read Full Article
25 Likes
Insider
4w
375
Image Credit: Insider
Stock market today: Indexes trade mixed with Nvidia's 3rd-quarter results on deck
- US stocks were mixed as traders looked ahead to Nvidia's earnings report, with Wall Street estimating $33 billion in revenue for the third quarter.
- Investors are eager to see if Nvidia's results can continue to show the strength of the artificial intelligence trade.
- Traders adjusted their expectations for Fed rate cuts after a slew of Fed speakers, with odds of a rate cut at the next policy meeting rising to 48%.
- Geopolitical tensions and rising hawkish comments from Fed officials added to market uncertainty.
Read Full Article
22 Likes
Medium
1M
40
Image Credit: Medium
My first time investing as a 19 years old: Lessons From Starting Small
- Starting to invest at a young age can make a huge difference.
- Starting with small amounts is sufficient to get started.
- Investing in index funds provides exposure to a range of companies.
- Invest in companies and sectors that you understand and believe in.
Read Full Article
2 Likes
Pymnts
1M
246
Image Credit: Pymnts
Wall Street Braces for Nvidia’s Trillion-Dollar Earnings Report
- Nvidia's earnings report could have a significant impact on its stock value.
- Uncertainty surrounds Nvidia's new Blackwell chip line due to production delays.
- Customers may delay purchases of current-generation Hopper chips while awaiting Blackwell's release.
- Nvidia's dominance in the AI industry and its potential for growth are highly regarded.
Read Full Article
14 Likes
Androidauthority
1M
138
Image Credit: Androidauthority
Pokémon Go trainers have actually been helping train AI, too
- AR mobile game maker Niantic is building a new type of AI model to understand and navigate physical spaces.
- The company is training the AI using data gathered from its mobile apps like Pokemon Go and Scaniverse.
- The AI could be used for supporting AR, robotics, content creation, and more.
- Niantic is well-known for creating popular AR games like Pokemon Go and apps like Scaniverse.
Read Full Article
8 Likes
Pcgamesn
1M
434
Image Credit: Pcgamesn
This free Nvidia GeForce RTX 3060 seems too good to be true, and it probably is
- A Reddit user claimed to have received a free Nvidia GeForce RTX 3060 from Chinese marketplace Temu.
- However, there are concerns that the card might be a fake or misleading GPU.
- Chinese e-commerce sites have been known to sell counterfeit graphics cards.
- While the user might have been lucky, the offer is likely too good to be true.
Read Full Article
26 Likes
For uninterrupted reading, download the app