According to the Gate News bot message, as per the official announcement from Gate on September 1, 2025: Gate will launch NodeOps( spot trading on September 1, 2025, at 19:00 ) UTC+8. NODE was previously launched on Gate Alpha on August 7, 2025. NodeOps Network is a platform that supports large-scale verifiable computing, providing GPU/CPU orchestration and RPC services. The project has secured $5 million in funding from institutions such as L1D, BFF, and Finality Capital, currently boasting over 700,000 users and managing assets exceeding $150 million. The platform connects more than 89,000 machines and over 24,000 suppliers, generating cumulative revenue exceeding $4.1 million, ranking among the top five in the revenue growth rate of DePIN projects.

Jin10 data reported on August 15, Lanke Technology announced the launch of the sixth-generation Jindai® performance core CPU today. This high-performance server processor, which integrates a groundbreaking architecture, full-stack compatibility, and chip-level security protection, aims to provide a powerful Computing Power engine for data centers, artificial intelligence, cloud computing, and critical industry infrastructure.

Jin10 data reported on August 14, the latest data from market research firm Jon Peddie Research shows that the global client CPU market has achieved growth for two consecutive quarters. In the second quarter of 2025, client CPU shipments increased by 7.9% quarter-on-quarter and 13% year-on-year; during the same period, server CPU shipments grew by 22% year-on-year, with a slight rise of 0.6% quarter-on-quarter.

According to Gate News bot, on July 25, 2025, Gate will launch the spot trading of Spheron Network (SPON) token for the first time and will start the 275th HODLer Airdrop activity. The airdrop event will end on July 29, 2025, at 15:00 ( UTC+8, with a total airdrop amount of 266,667 SPO. The SPO/USDT trading pair will start trading on July 29, 2025, at 17:00 ) UTC+8. Spheron Network is a decentralized computing network that integrates data center-grade and retail-grade GPU/CPU resources, with over 50,000 active nodes. The SPON Token is issued on the BASE network, with a total supply of 1 billion tokens, and the contract address is 0x080d43c2164afdbc3712422ce78ab902ccab5ca1.

Researchers have found that malicious Monero Mining software has once again targeted 3,500 websites, employing covert techniques to limit CPU usage and implement a "low-key, slow mining" strategy to avoid detection.

Gate News Bot news, July 21, according to CoinMarketCap data, as of the time of writing, MBC (MicroBitcoin) is currently priced at 0.00024538 USD, having risen 65.27% in the last 24 hours, reaching a high of 0.00045029 USD and a low of 0.0001251 USD. The 24-hour trading volume reached 605,500 USD. The current market capitalization is approximately 13.5 million USD, an increase of 5.35 million USD compared to yesterday. MicroBitcoin is a payment protocol built on blockchain technology, centered around open source, decentralization, and peer-to-peer. MBC aims to become a practical payment method with features such as a 1-minute block time, low transaction fees, and fast confirmations. The project uses the Power2b algorithm to achieve CPU-only mining and employs Bl

Gate News bot message, Allnodes has launched a bare-metal server service specifically designed for Solana validators and developers, providing high-performance unmanaged storage solutions. The system features AMD EPYC Turin CPU (24 to 64 cores), 256GB to 3072GB DDR5 ECC RAM, and NVMe Gen 5 SSD with read speeds up to 14500 MB/s. The server supports infrastructure such as Jito-Solana and advanced MEV strategies.

Gate News bot message, cybersecurity company Kaspersky reported a mining attack incident targeting Russian industrial equipment. The hacker organization Librarian Ghouls (also known as Rare Werewolf) has been launching phishing email attacks against Russian industrial enterprises and engineering schools since December 2024, affecting Belarus and Kazakhstan as well. The attackers infiltrated hundreds of devices through disguised official documents. The organization remotely disabled the security systems of the target devices between 1 AM and 5 AM, utilizing the devices' RAM, CPU, and GPU to conduct cryptocurrency Mining. According to a Kaspersky investigation, the organization used legitimate third-party software instead of homemade malware to carry out the attack, displaying characteristics of hacker activism. Source: Cointelegraph.

Golden Finance reports that the hacker group Librarian Ghouls (also known as Rare Werewolf) has infiltrated hundreds of Russian devices and is using them for crypto mining. The group spreads malware through phishing emails disguised as legitimate organizations, establishing remote connections after infecting the devices and disabling security systems such as Windows Defender. Hackers collect information on the devices' RAM, CPU cores, and GPU to optimize the configuration of the crypto mining programs. This hacking incident began in December 2023, with the attack primarily affecting industrial enterprises and engineering schools in Russia, while there were also victims in Belarus and Kazakhstan. Kaspersky speculates that Librarian Ghouls may be hacktivists, as they rely on legitimate third-party tools rather than developing their own malware.

NVIDIA CEO Jen-Hsun Huang announced the launch of NVLink Fusion, which supports multiple hardware and aims to strengthen the construction of enterprise AI infrastructure, with customizable configurations for CPU, GPU, and ASIC, addressing the complexity challenges of building enterprise AI environments.

According to Gate.io News bot, RISC-V, an open-source CPU instruction set architecture originating from the University of California, Berkeley, has garnered significant attention in the tech field for its clarity and simplicity. CKB-VM has chosen to apply this traditional instruction set architecture, used for CPU verification, in the development of blockchain virtual machines. This technological innovation not only achieves a deep integration of the hardware and software industries but also brings a more open development environment to the blockchain ecosystem. Source: wublock123

Jin10 data reported on April 14th, Intel's core design team senior chief engineer Ori Lempel stated in an interview that the company's CPU architecture has decoupled 99% from the process Node, which gives the product team greater process flexibility. Specifically regarding the progress, Intel's CPU core architecture team is currently engaged in the core architecture design work for the third generation after Panther Lake.

The Loongson 2K3000 (3B6000M) has performed well in preliminary functionality and performance tests, marking that Loongson chips have mastered the key core technologies of general-purpose processors, graphics processors, and AI processors. The Loongson processor has entered a new phase, focusing on developing AI processors to provide high-performance and cost-effective CPU chip products for various fields.

NVIDIA unveiled the next generation Vera Rubin AI superchip and Blackwell Ultra at the GTC conference. Vera Rubin is set to launch in the second half of 2026, with performance twice that of the previous generation; equipped with 288GB of HBM4. The next generation Vera Rubin Ultra will be launched in 2027, combining CPU and GPU, with Rubin Ultra consisting of four GPUs.

Loongson Technology stated that industry customers have a greater demand for servers than government customers. They have achieved some success with the 5000 series server CPU and plan to launch the 3C6000 series, hoping to continue their success. The market layout will focus on "point" applications, such as the storage server field.

The smart ZK engine Brevis releases high-performance modular general zkVM Pico. Developers can use Rust language and ZK coprocessor for optimization, improving application performance and user experience. Pico performs excellently in the latest CPU performance Benchmark tests and will release a GPU accelerated version in the future.

Loongson successfully launched the DeepSeek R1 7B model with the Loongson 3 CPU, achieving localized deployment with outstanding performance and excellent cost, providing users with a faster, stronger, and more efficient training and inference experience. Loongson is collaborating with partners such as Taisui Yuanyi, Cambricon, TianSuan Zhixin, AlgoTech, openEuler to build a multi-form inference platform for the DeepSeek series models.

On January 26, Jinshi Data reported that on January 26, Lingyi Zhizao announced that it has been promoted to be a core supplier of AMD and will cooperate in the field of heat dissipation in AMD's different series of products. In terms of thermal management, there are cooperative opportunities for heat dissipation products in various fields such as GPU, CPU, and AI applications (servers, notebooks, PCs, graphics cards, wearable devices, humanoid robots).

Jinshi data news on January 22nd, Arm is preparing to significantly increase the licensing fees (up to 300% increase), which will have a serious impact on the future development of Samsung Exynos chips. Samsung once had its own team to develop custom CPU cores, but due to unsatisfactory power consumption and performance, Samsung disbanded the department in November 2019 and decided to adopt Arm's public design.

Golden Ten Data reported on December 16 that Loongson Zhongke's recent research minutes show that the company's next-generation desktop chip 3B6600 is currently in the design stage and is expected to be delivered in the first half of next year. The desktop chip 3B6600 is an 8-core desktop CPU with integrated GPGPU and PCIE interfaces.

CITIC Securities research report believes that there will be a recovery opportunity for the global enterprise IT hardware zone by 2025. It is expected that the macro situation will improve by 2025, the policy uncertainty after the US election will be eliminated, the Federal Reserve will drop interest rates, and it will help to ease overall IT spending by European and American enterprises. At the same time, factors such as the end of Windows 10 EOL, the introduction of new generation server CPU platforms, and the end of inventory digestion of enterprise network equipment will also form support in terms of sales and structure. It is recommended that investors follow the commercial PC, AI server, general server, network equipment, and enterprise storage zone.

Odaily Planet Daily News ETH Staking Infrastructure SSV Network official announced on platform X that it will upgrade Alan on Mainnet at Block Height 327,375, which is expected to take place at 20:00 on November 25th. This upgrade will bring various performance and scalability improvements, significantly dropping CPU usage.

On November 20th, according to a report by Jon Peddie Research, the overall PC CPU market rose by 12.2% in the third quarter of 2024 compared to the second quarter of 2024. In the second quarter, PC

Jingchen shares released performance meeting, introducing the new generation 6nm chip based on ARMV9 architecture and self-developed edge AI capability. This chip improves CPU performance by more than 60%, GPU performance by more than 230%, and reduces power consumption by 50%. In addition, the chip also integrates multiple functions such as edge AI, subtitle translation, and real-time meeting recording. It is the industry's first 6nm commercial chip that integrates 4K and AI functions.

Golden Ten Data Nov. 15th news, Lingyi Intelligent Manufacturing stated in the investor relations activity record that the company has obtained the supplier code from Supermicro Semiconductor Company for cooperation opportunities in the field of heat dissipation products for GPU, CPU and AI applications (servers, notebooks, PCs, graphics cards, wearable devices, humanoid robots), and future orders are expected to be gradually released.

On October 11th, Jinshi Data reported that Runhe Software stated during a recent survey that the company continues to increase its research and development investment in internet of things basic software, such as operating systems, cloud computing platforms, big data platforms, and artificial intelligence components. HopeOS

Odaily Planet Daily News BTC L2 network Stacks announced in a post on X that the upcoming Nakamoto Release will utilize the CPU power of BTC to achieve 100% BTC Finality Settlement Stacks transactions. Nakamoto RC2

Ethereum has introduced the EIP-7732 specification, which has achieved Enshrined Proposer-Builder Separation, separating execution validation from consensus validation, while PBS is directly embedded in Ethereum's consensus layer. The specification modifies aspects such as beacon chain, fork choice rule, and P2P communication, supporting validators to act as builders and providing trustless exchange, enabling validators to effectively utilize CPU. In addition, the specification also includes a timely payload committee, with a subset of the validation committee voting on the timeliness of the builder's payload release, making it compatible with future upgrades.

On June 26th, io.net announced the launch of block rewards, which provide block rewards to Suppliers accounts that provide GPU or CPU on the IOG network. Block rewards are credited to Suppliers accounts in IO form on an hourly basis. The current distribution of block rewards is: 80% released per hour allocated to GPU and 20% allocated to CPU.

Jinshi data news on June 24th, Longxin Zhongke's recent research notes show that Longxin 3B6600 integrates eight CPU cores, dual-channel DDR5, PCIE4, GPGPU, and strives to increase the single-core frequency performance by 20%-30% compared to 3A6000, and plans to mass produce in 2025H1.

The Base team will increase the gas target for each block to 8.125Mgas/second to scale up, making it fast, cheap, and easy to use. For most users, no changes are needed, but users running nodes need to monitor CPU and disk I/O usage, enhance disk throughput, use disks with lower read/write latency, improve CPU, and focus on single-core speed.

In a paper published Thursday, academic researchers revealed that a newly discovered vulnerability in Apple's M-series chips could cause a leak of wallet keys. The vulnerability is a side channel that allows end-to-end key extraction when Apple silicon runs a crypto wallet, but this vulnerability cannot be patched directly because it stems from the microarchitecture design of the chip itself. Instead, the problem can only be mitigated by building defenses into third-party encryption software that can significantly degrade the performance of the M-series when performing cryptographic operations, especially on earlier M1 and M2 generations. This vulnerability can be exploited when the target cryptographic operation and a malicious application with normal user system privileges are running on the same CPU cluster.

Odaily Planet Daily News Nvidia founder Jensen Huang delivered an opening speech at the GTC conference held in San Jose, USA. In this presentation, NVIDIA unveiled their next-generation chip architecture, Blackwell. According to reports, the Blackwell GPU is named after mathematician David Harold Blackwell, and it also follows the Hopper architecture previously launched by NVIDIA. With 208 billion transistors in a Blackwell GPU, it can support AI models with up to 10 trillion parameters. In addition to the chip itself, this architecture also uses the fifth-generation NVLink high-speed interconnect and the second-generation Transformer engine, which is comprehensively upgraded in many aspects. According to Huang, the new chip will be available later in 2024. Huang said that in the future, Nvidia plans to use Blackwell to expand into AI companies around the world, signing contracts with all OEMs, regional clouds, national sovereign AI, and telecommunications companies around the world. At present, Amazon, Dell, Google, Meta, Microsoft, OpenAI, and Tesla have all planned to use Blackwell GPUs. It's worth noting that the fabled B100 didn't show up before, and Nvidia instead released a superchip GB, which is a combination of a Grace CPU and two sets of Blackwell GPUs

UnionTech Software officially announced that UOS AI has been officially launched on the UnionTech UOS desktop platform yesterday, which is compatible with cloud-side and device-side large models, and supports access to local models deployed privately. According to the official introduction, after installing UOS AI, users can wake up the UOS AI desktop intelligent assistant through the shortcut key "super+C" to conduct voice conversations or natural language interactions. According to reports, at the application layer, the desktop intelligent assistant supports multi-modal input and output, and supports more than 40 use scenarios of natural language command invocation, and currently has more than 10 applications connected to UOS AI, covering system file management (global search, picture viewing, photo album, etc.), office efficiency tools (browser, mailbox, free writing, etc.), productivity tools (deepin-IDE, etc.); at the chip layer, UOS AI supports domestic mainstream CPU chips and NVIDIA, Moore Threads and other mainstream GPUs at home and abroad Chip; At the large model layer, UOS AI covers 90% of mainstream open source large models and AI frameworks, and supports cloud and local hybrid models.