GPU
-
OpenAI launches GPT-6 Astra Ultrafast, accelerated by NVIDIA Blackwell GPUs, boosting generation speed by 8x.
OpenAI announced the launch of the GPT-6 Astra Ultrafast model, now available via the OpenAI API and for eligible ChatGPT Work and Codex users. The Ultrafast mode runs on NVIDIA Blackwell GPUs and off
-
NVIDIA Details How Its AI Factories Maximize ROI, Citing 30x Throughput Increase
NVIDIA's official blog explains that AI factories, costing roughly $60 million per megawatt, maximize return on investment through three key factors: productivity, durability, and fungibility. The com
-
Wells Fargo: AWS to raise GPU compute service prices by 15% again on October 7, Amazon maintains "Overweight" rating
Wells Fargo reports that Amazon Web Services (AWS) will raise GPU computing service prices by another 15% on October 7, following a 20% increase in July. This marks the fourth consecutive quarter AWS
-
NVIDIA (NVDA) Encyclopedia: Deconstructing the AI Computing Leader's Moat and Valuation Logic, from GPU Hardware to the CUDA Software Ecosystem
As of September 2026, NVIDIA has become the world's most valuable company by market capitalization, with its AI hardware and CUDA software ecosystem forming a formidable competitive barrier. The company continues to launch new-generation GPU architectures such as Blackwell and Vera Rubin, and is expanding the CUDA platform to support Rust language and quantum computing. NVIDIA's valuation is closely tied to the global AI capital expenditure cycle. Despite competition from AMD, Intel, and cloud service providers' self-developed chips, its data center revenue and profitability remain strong. However, the market still has concerns regarding its high valuation, geopolitical risks, and founder succession risk.
-
Sharon AI signs five-year agreement with Rafay Systems to orchestrate up to 150,000 GPUs
Australian neocloud Sharon AI, a Nasdaq-listed company, has signed a five-year agreement with Rafay Systems to use its platform as a centralized orchestration layer across its AI Factory environments.
-
Moore Threads, a leading Chinese GPU manufacturer, hit the 20% daily limit down on Monday as 25.77 million restricted shares were unlocked, nearly doubling its float.
Moore Threads, a leading domestic GPU manufacturer, saw its stock price hit the 20% daily limit down on September 7, closing at 415.48 yuan per share, with its total market capitalization falling to 1
-
Broadcom CEO Hock Tan: Expects AI revenue to double consecutively over the next two years, reaching $230 billion in fiscal year 2028; custom chips to match GPU performance at half the cost.
After market close on Wednesday ET, Broadcom announced its net revenue for the third fiscal quarter of 2026 surged 86% year-over-year to $29.591 billion, with AI semiconductor revenue soaring 221% yea
-
Ming-Chi Kuo: NVIDIA has restarted the Rubin CPX AI inference pre-fill optimization chip project, with hashrate close to the standard Rubin GPU, expected to enter production in Q1 2027.
According to a new industry survey by renowned analyst Ming-Chi Kuo, NVIDIA has restarted the "Rubin CPX" AI inference prefill acceleration GPU project and significantly adjusted its product design, t
-
Nvidia is acting as a matchmaker to connect GPU buyers with data center operators in the Nordics, aiming to accelerate AI infrastructure deployment and address GPU deployment bottlenecks.
The company's CFO Colette Kress confirmed in June that Nvidia has engaged in such matchmaking, helping customers secure land, power, and shell infrastructure. The Nordics, particularly Oslo, Stockholm
-
24/7 Wall St. gives AMD a "high-conviction buy" rating with a target price of $629.60, implying a 31% upside.
The agency noted that AMD's second-quarter data center revenue surged 107% to $6.7 billion, with total revenue reaching $11.54 billion, a year-over-year increase of 50.11%. The company's non-GAAP earn
-
RUM Group stock rallies 7% on $13.7 billion GPU supply contract, but company admits it lacks funds to fulfill the deal
RUM Group (NASDAQ:RUM) stock surged 7% to $9.68 after announcing a commercial agreement worth approximately $13.7 billion to supply GPUs and GPU services to an unnamed U.S.-based cloud customer. The c
-
CoreWeave CEO Says Company Is Booking NVIDIA GPUs at "Full Price" Through 2029
CoreWeave CEO Mike Intrator stated in an interview with CNBC on Wednesday, August 12, that the AI infrastructure market is sending strong signals about demand: even older NVIDIA GPUs, which investors
-
NVIDIA CEO Jensen Huang will update the trillion-dollar GPU business outlook on August 26, and expects CPU revenue to reach $20 billion this year, with a potential market of $200 billion.
NVIDIA CEO Jensen Huang will provide an update on his previously projected $1 trillion GPU (graphics processing unit) business opportunity after the company's Q2 FY2027 earnings release on August 26,
-
NVIDIA Chips Face 12:1 Supply-Demand Ratio, Experts Say, as Meta, Oracle, and Others Turn to Hyperscale Computing, Intensifying GPU Demand
Dan Ives, an analyst at Wedbush Securities, noted that NVIDIA (Nvidia) chips face a 12:1 supply-demand ratio, thanks to companies like Meta and Oracle actively shifting to hyperscale computing, which
-
Axe Compute Secures Over $1.3 Billion in AI Infrastructure Contracts; ARR Expected to Exceed $384 Million Following Deployment
Svmuu News: According to an official announcement, Axe Compute (NASDAQ: AGPU), a publicly traded AI infrastructure company listed on the Nasdaq, announced that it has secured more than $1.3 billion in
-
Wang Dong, Co-founder of Moore Threads: There Is No “One-Size-Fits-All Chip” in the Inference Market—Instead, It’s a Combination of Solutions
Svmuu News: Wang Dong, co-founder and CEO of GPU manufacturer Moore Threads, said, “ Large-scale models are developing very rapidly both domestically and internationally. Currently, leading companies
-
The South Korean government is considering extending its “Sovereign AI Core GPU Security” program, with a budget of up to approximately 4 trillion won for next year.
Svmuu News: The South Korean government is considering extending its support program to ensure access to GPUs, a core resource for AI. On July 7, according to industry sources, the Ministry of Science
-
Micron (MU) is more important than NVIDIA! Citrini analyst says AI model inference performance depends more on memory than on GPUs
Svmuu News: In response to the community debate over “Micron vs. NVIDIA,” Citrini Research analyst Jukan posted on X, stating, “Although MU (Micron) is not Nvidia, its future significance may surpass
-
OpenAI launches GPT-6 Astra Ultrafast, accelerated by NVIDIA Blackwell GPUs, boosting generation speed by 8x.
OpenAI announced the launch of the GPT-6 Astra Ultrafast model, now available via the OpenAI API and for eligible ChatGPT Work and Codex users. The Ultrafast mode runs on NVIDIA Blackwell GPUs and off
-
NVIDIA Details How Its AI Factories Maximize ROI, Citing 30x Throughput Increase
NVIDIA's official blog explains that AI factories, costing roughly $60 million per megawatt, maximize return on investment through three key factors: productivity, durability, and fungibility. The com
-
Wells Fargo: AWS to raise GPU compute service prices by 15% again on October 7, Amazon maintains "Overweight" rating
Wells Fargo reports that Amazon Web Services (AWS) will raise GPU computing service prices by another 15% on October 7, following a 20% increase in July. This marks the fourth consecutive quarter AWS
-
Sharon AI signs five-year agreement with Rafay Systems to orchestrate up to 150,000 GPUs
Australian neocloud Sharon AI, a Nasdaq-listed company, has signed a five-year agreement with Rafay Systems to use its platform as a centralized orchestration layer across its AI Factory environments.
-
Moore Threads, a leading Chinese GPU manufacturer, hit the 20% daily limit down on Monday as 25.77 million restricted shares were unlocked, nearly doubling its float.
Moore Threads, a leading domestic GPU manufacturer, saw its stock price hit the 20% daily limit down on September 7, closing at 415.48 yuan per share, with its total market capitalization falling to 1
-
Broadcom CEO Hock Tan: Expects AI revenue to double consecutively over the next two years, reaching $230 billion in fiscal year 2028; custom chips to match GPU performance at half the cost.
After market close on Wednesday ET, Broadcom announced its net revenue for the third fiscal quarter of 2026 surged 86% year-over-year to $29.591 billion, with AI semiconductor revenue soaring 221% yea
-
Ming-Chi Kuo: NVIDIA has restarted the Rubin CPX AI inference pre-fill optimization chip project, with hashrate close to the standard Rubin GPU, expected to enter production in Q1 2027.
According to a new industry survey by renowned analyst Ming-Chi Kuo, NVIDIA has restarted the "Rubin CPX" AI inference prefill acceleration GPU project and significantly adjusted its product design, t
-
Nvidia is acting as a matchmaker to connect GPU buyers with data center operators in the Nordics, aiming to accelerate AI infrastructure deployment and address GPU deployment bottlenecks.
The company's CFO Colette Kress confirmed in June that Nvidia has engaged in such matchmaking, helping customers secure land, power, and shell infrastructure. The Nordics, particularly Oslo, Stockholm
-
24/7 Wall St. gives AMD a "high-conviction buy" rating with a target price of $629.60, implying a 31% upside.
The agency noted that AMD's second-quarter data center revenue surged 107% to $6.7 billion, with total revenue reaching $11.54 billion, a year-over-year increase of 50.11%. The company's non-GAAP earn
-
RUM Group stock rallies 7% on $13.7 billion GPU supply contract, but company admits it lacks funds to fulfill the deal
RUM Group (NASDAQ:RUM) stock surged 7% to $9.68 after announcing a commercial agreement worth approximately $13.7 billion to supply GPUs and GPU services to an unnamed U.S.-based cloud customer. The c
-
CoreWeave CEO Says Company Is Booking NVIDIA GPUs at "Full Price" Through 2029
CoreWeave CEO Mike Intrator stated in an interview with CNBC on Wednesday, August 12, that the AI infrastructure market is sending strong signals about demand: even older NVIDIA GPUs, which investors
-
NVIDIA CEO Jensen Huang will update the trillion-dollar GPU business outlook on August 26, and expects CPU revenue to reach $20 billion this year, with a potential market of $200 billion.
NVIDIA CEO Jensen Huang will provide an update on his previously projected $1 trillion GPU (graphics processing unit) business opportunity after the company's Q2 FY2027 earnings release on August 26,
-
NVIDIA Chips Face 12:1 Supply-Demand Ratio, Experts Say, as Meta, Oracle, and Others Turn to Hyperscale Computing, Intensifying GPU Demand
Dan Ives, an analyst at Wedbush Securities, noted that NVIDIA (Nvidia) chips face a 12:1 supply-demand ratio, thanks to companies like Meta and Oracle actively shifting to hyperscale computing, which
-
Axe Compute Secures Over $1.3 Billion in AI Infrastructure Contracts; ARR Expected to Exceed $384 Million Following Deployment
Svmuu News: According to an official announcement, Axe Compute (NASDAQ: AGPU), a publicly traded AI infrastructure company listed on the Nasdaq, announced that it has secured more than $1.3 billion in
-
Wang Dong, Co-founder of Moore Threads: There Is No “One-Size-Fits-All Chip” in the Inference Market—Instead, It’s a Combination of Solutions
Svmuu News: Wang Dong, co-founder and CEO of GPU manufacturer Moore Threads, said, “ Large-scale models are developing very rapidly both domestically and internationally. Currently, leading companies
-
The South Korean government is considering extending its “Sovereign AI Core GPU Security” program, with a budget of up to approximately 4 trillion won for next year.
Svmuu News: The South Korean government is considering extending its support program to ensure access to GPUs, a core resource for AI. On July 7, according to industry sources, the Ministry of Science
-
Micron (MU) is more important than NVIDIA! Citrini analyst says AI model inference performance depends more on memory than on GPUs
Svmuu News: In response to the community debate over “Micron vs. NVIDIA,” Citrini Research analyst Jukan posted on X, stating, “Although MU (Micron) is not Nvidia, its future significance may surpass
-
NVIDIA (NVDA) Encyclopedia: Deconstructing the AI Computing Leader's Moat and Valuation Logic, from GPU Hardware to the CUDA Software Ecosystem
As of September 2026, NVIDIA has become the world's most valuable company by market capitalization, with its AI hardware and CUDA software ecosystem forming a formidable competitive barrier. The company continues to launch new-generation GPU architectures such as Blackwell and Vera Rubin, and is expanding the CUDA platform to support Rust language and quantum computing. NVIDIA's valuation is closely tied to the global AI capital expenditure cycle. Despite competition from AMD, Intel, and cloud service providers' self-developed chips, its data center revenue and profitability remain strong. However, the market still has concerns regarding its high valuation, geopolitical risks, and founder succession risk.
GPU
24H Trending
-
1
Oasys (OAS) Project Analysis: The Intersection of Gaming Chains and On-Chain Perpetual Contracts
-
2
Global Virtual Currency Wallet Types and the Current State of Cryptocurrency Trading Platforms in Mainland China
-
3
HAN Token Analysis: What is HanChain? How to Trade and Acquire It?
-
4
RSS3: Web3 Decentralized Information Protocol and Its Token Analysis
-
5
Crypto market enters October after strong Q3, eyes Ethereum testnet and ETF flows
-
6
ENJ Trading Guide: How to Buy and Sell Enjin Tokens and Their Listed Exchanges
-
7
YFC (Yearnchange) Token Analysis: The Current State and Risks of DeFi Copycats
-
8
Aave Official Community Channels Explained: Do WeChat Groups Exist and How to Participate Safely
-
9
An Analysis of USDC's Performance and Role During Financial Crises
-
10
BAI Token Analysis: An Overview of a Crypto Asset with Multiple Identities
Markets Today
Recommended Reading










