Early morning on November 11 Beijing time, CoreWeave released its Q3 FY2026 earnings report. Revenue surged nearly 160% year-over-year to approximately $3.5 billion, yet adjusted loss per share significantly exceeded expectations. What unsettled the market more than the numbers themselves was the industry anxiety condensed within this NVIDIA-aligned cloud provider: Is the rapid surge in AI computing power built on a solid foundation of revenue, or is it a castle in the air constructed with debt?
November 11, 2026
Iluvatar CoreX showed how domestic AI chips move from benchmarks to cloud-edge deployment. Its Tiangai 300 cloud GPU claims near-H200 throughput, while software Agents, Skills and FlyDSL shorten deployment, tuning and compilation. A heterogeneous AFD scheme cuts decode latency 60%. The Tongyang edge lineup, led by the 300 TOPS TY1200, delivers high-performance offline AI, local video generation, safety-sensitive processing and physical simulation; TY2000 will reach 600T.
November 9, 2026
Author: Fang Yuan If an enterprise prepares to deploy 1,000 agents, what else does it need besides purchasing GPUs? The answer may be more complex than imagined. Each agent requires an execution environment, memory to save running states, storage to record context, and a CPU to coordinate tasks and invoke tools. As the number of agents increases, resource consumption that was previously negligible can become a major expense for data centers. This reflects a shift occurring in the AI industry. Ov
November 4, 2026
The volume of AI bubble skepticism reached a peak in recent weeks. Critics point to the same issue: massive compute investments have been deployed, but is anyone actually using them? Has demand been overextended? AMD's earnings report states a more straightforward fact: customers with identifiable names are buying its AI chips with real money, and purchasing more than they did last quarter. Data Center: Doubled Year-over-Year, Still Accelerating Quarter-over-Quarter AMD's revenue for Q3 2026 was
November 4, 2026
As Artificial Intelligence (AI) moves from cloud-based large models to physical-world applications such as automobiles and robots, the competitive logic for edge-side AI chips is evolving. For chips, AI computing power figures remain important but are no longer sufficient on their own to determine competitiveness. Taking intelligent driving as an example, whether large models can run efficiently on the vehicle side, whether chips can keep pace with the rapid evolution of model requirements, whet
November 2, 2026
On October 19, the 2026 RISC-V China Summit was held in Shenzhen. Chen Zhijian, Founder and CEO of domestic RISC-V chipmaker SpacemiT, officially unveiled the company's first server-grade RISC-V chip, the V100, at the summit. Integrating 46 RISC-V cores and supporting 8-channel DDR5 and 52 PCIe 5.0 lanes, this chip is hailed as the "first computing chip to fully comply with RISC-V server specifications." It marks SpacemiT's official transition from the edge to the cloud, and signifies that RISC-
October 21, 2026
At the recently held 26th China International Industry Fair, D-Robotics made its debut exhibition. Unlike traditional chip vendors that showcase products independently, D-Robotics dedicated more than half of its booth space to ecosystem partners. Source: D-Robotics During the fair, Jiang Zhou, Vice President of Sales at D-Robotics, was interviewed by EEFOCUS reporters, elaborating on the strategic thinking and implementation paths of this AI chip company, which spun off from Horizon Robotics, ac
October 21, 2026
As "AI bubble" becomes a topic of widespread discussion, Montage Technology released a Q3 earnings preview showing net profit growth exceeding 150% year-over-year. Setting aside stock price fluctuations and valuation levels, the true value of this report lies in its provision of technical testimony regarding the authenticity of AI demand from a critical niche within the supply chain. The market's fear of an AI bubble essentially stems from skepticism about whether "computing power accumulation c
October 21, 2026
On the evening of October 13, SigmaStar Technology released its performance forecast for the first three quarters of 2026. Net profit attributable to shareholders is expected to range from CNY 1.543 billion to CNY 1.613 billion, representing a year-on-year increase of 663% to 698%. With net profit projected to surge by nearly seven times at the upper end, the financial narrative for edge-side SoCs (System on Chip) is being rewritten. A closer look reveals the trend more clearly. Net profit for t
October 14, 2026
Recently, Annie Shea Weckesser, Chief Marketing and Communications Officer of Intel, delivered a keynote speech titled "Intelligence Ushers in the AI Era" at the 2026 Intel Technology Innovation and Industry Ecosystem Conference. In her speech, Annie Shea Weckesser introduced Intel's three-pronged strategy to collaborate with the industry ecosystem to drive the implementation of intelligence: first, building an open computing platform covering PCs, edge, and data centers; second, designing custo
October 5, 2026
By Fang Yuan "10,000 sandboxes spun up within 10 minutes." During Intel Connection 2026, Intel put forward this figure. The creation time for each sandbox must not exceed 60 milliseconds. A sandbox is not an empty shell; it must host a complete runtime environment, including images, file systems, and isolated spaces. This requirement is not a laboratory hypothesis; it comes from the production environments of real customers. It also points to a specific question: as agents evolve from chatting t
September 29, 2026
Author | Yunpeng, Editor | Moying From the cloud to the edge, today AI is profoundly changing the way humans interact with the physical world. From medical wearable devices and smart homes to industrial automation and embodied robots, edge AI, with its unique advantages of low latency, high security, and high reliability, has brought new solutions for AI deployment in many industries. Massive AI applications directly complete real-time perception, inference, and decision-making locally on termin
September 27, 2026
Over the past two years, the computing power market has been almost entirely dominated by GPUs. When training peaks could not be pushed higher, all other efforts seemed secondary, and the landscape was paced by a few accelerator cards. However, as the industry chain moves past its most boisterous phase, the trend is shifting from "making models larger" to "making models usable." Attention is quietly moving away from the peak computing power of a single chip to whether the entire system can handl
September 24, 2026
The capacity tightness driven by AI computing demand is steadily transmitting to the pricing of wafer foundry services. According to Digitimes, citing supply chain sources, TSMC plans to raise its wafer foundry quotes starting from January 2027, with an overall price increase ranging from 3%–6%. Advanced nodes such as 2nm and 3nm will see higher increases; mature and specialty processes will not undergo a uniform price adjustment, but will be negotiated individually based on customers and capaci
September 24, 2026
Author | Yunpeng Editor | Moying As AI accelerates its shift from the cloud to the edge and endpoint, Agentic AI is rapidly penetrating various industries. A series of physical industries, including transportation, oil and gas, robotics, healthcare, and manufacturing, have been swept into the wave of intelligent transformation. Computing power support at the edge and endpoint has become a critical cornerstone driving the digital and intelligent transformation of the physical world. Such changes
September 23, 2026
Author: Peng Cheng Over the past few years, computing power has been primarily concentrated in the cloud, such as data centers. However, as industry intelligence continues to deepen, an increasing number of computing tasks are shifting to the edge and on-site devices. There are three main reasons for this shift in computing power. First is real-time performance: industrial control requires microsecond-level response and multi-task concurrency, making a control loop that uploads data to the cloud
September 23, 2026
Abstract: As 800G, 1.6T, and even CPO evolve rapidly, the optical module MCU is upgrading from a basic controller to a critical management chip for complex optoelectronic systems. GigaDevice is also leveraging its dedicated MCUs, Flash, analog capabilities, and supply chain synergy to further strengthen its competitiveness in the AI optical interconnect market. In recent years, the rapid expansion of AI computing infrastructure has been transforming data transmission within data centers. 800G op
September 18, 2026
ResearchInChina and Zuosi Auto Research jointly released the "2026 Automotive Microcontroller (MCU) and Application Scenario Research Report". In the vehicle's electrical and electronic (E/E) architecture, MCUs are widely distributed across various functional domains of the automobile, undertaking tasks such as functional safety monitoring, real-time control, sensor preprocessing, communication power management, and redundant supervision. Functional safety levels range from ASIL A to ASIL D depe
September 17, 2026
In September 2026, the mobile chip market is set for an unprecedented fierce battle. The Apple A20 Pro, MediaTek Dimensity 9600 Pro, Huawei Kirin 9050 Pro, and the recently launched Xiaomi Xuanjie O3 in August—these four chips are colliding head-on within the same timeframe. Interestingly, they have adopted four distinctly different technological approaches. Apple A20 Pro On the Geekbench benchmark website, the A20 Pro in the iPhone 18 Pro Max scored 4,727 in single-core and 12,424 in multi-core
September 16, 2026
Abstract: AI (Artificial Intelligence) data centers are driving the rapid evolution of optical modules from 400G and 800G to 1.6T, and a long-overlooked MCU (Microcontroller Unit) is evolving accordingly. Although it does not participate in high-speed data transmission, it handles tasks such as the CMIS protocol, firmware management, analog monitoring, fault protection, and device coordination. With the continuous addition of Flash, I3C, analog resources, and security features, optical module MC
September 15, 2026