AI Chips news and analysis on Web Pulse: 3106 curated articles tracking the latest AI Chips developments, tools, and research, updated continuously from vetted sources.
At Computex, Delos Data demonstrated a modular server chassis designed to help AI chip startups deploy rack-scale systems more quickly. The front of the system contains 36 OSFP cages rated at 1.6 Tbps each, providing app…
Kioxia Holdings surpassed Toyota Motor to become Japan’s largest company by market value, with shares surging 7.6% Friday to lift its market capitalization above ¥44 trillion ($274 billion). The memory chipmaker’s rise, …
Delos Data launched a service that helps AI chip startups build rack-scale systems by handling networking and integration challenges, reducing the complexity of competing with Nvidia NVL or AMD Helios platforms. The offe…
Kioxia Holdings Corp. briefly surpassed Toyota Motor Corp. as Japan’s most valuable company on June 3, reaching an intraday market capitalization of approximately ¥45 trillion ($281 billion). The memory chipmaker’s stock…
A developer benchmarked the Mac Studio M4 Max against the Mac Mini M4 Pro for local AI inference, finding that the Studio roughly doubles token generation speed on 70B models due to its 546 GB/s memory bandwidth versus t…
A developer found that running a 70B parameter LLM locally with only 8GB of VRAM is possible but requires significant optimizations and trade-offs. While a 70B model in FP16 format demands up to 140GB of VRAM, 4-bit quan…
Two ESP32-S3 microcontrollers running a Llama-architecture language model have achieved the first multi-chip pipelined LLM inference on ESP32-class hardware, splitting layers across two boards connected by three jumper w…
Global prime brokers are reducing margin financing for hedge funds betting on Samsung Electronics and SK Hynix, as the two stocks now represent roughly 40% of South Korea's Kospi Index. The pullback follows a blistering …
Researchers introduced SupraSNN, a hardware-software co-design that treats synaptic events as parallelizable micro-operations and physically decouples synaptic and neuronal computation, achieving synapse-level parallelis…
Vivo will launch the X Fold 6 with a custom Dimensity 9500 Super Edition processor co-developed with MediaTek, claiming the chip boosts peak AI processing performance by 111 percent while reducing power consumption by 56…
China plans to invest approximately $295 billion over the next five years to build a nationwide network of AI-focused data centers, with a target completion date of 2028. The initiative, led by the National Development a…
KKR is exploring strategies to overcome a major bottleneck in AI infrastructure, potentially involving significant investments in data centers and energy resources. The move highlights the growing demand for computing po…
Researchers at an undisclosed institution reverse-engineered Apple's Metal 4.1 tensor compute path on the M4 Max GPU, revealing that the fp8 matmul2d operation is emulated rather than hardware-accelerated. The study, pub…
Blaize announced its edge AI strategy, a partnership with Nokia, and a revenue outlook of $130 million for 2026, positioning itself in the competitive AI chip market.
South Korea's Hanmi Semiconductor announced Friday it plans to acquire 50 billion won ($32.9 million) in SpaceX shares ahead of the U.S. space company's stock market debut. The investment, representing about 7.24 percent…
Google is in advanced talks with Samsung Electronics to manufacture a component of its tenth-generation Tensor Processing Unit, codenamed Icefish, under a split-manufacturing plan that would also involve TSMC and MediaTe…
Apple announced five third-generation Apple Foundation Models during WWDC26, including on-device models AFM 3 Core and AFM 3 Core Advanced, and server-based models AFM Cloud, ADM 3 Cloud (Image), and AFM 3 Cloud Pro. For…
The Nasdaq-100 will add CoreWeave, Rocket Lab, Astera Labs, Nebius Group, and Teradyne to its flagship index effective June 22, replacing Charter Communications, Cognizant, Insmed, Verisk Analytics, and Zscaler. The quar…
UBS has backed Cerebras (CBRS) in the AI inference race, signaling growing investor confidence in the chipmaker's technology. The investment positions Cerebras as a key competitor in the AI infrastructure market.
Modal AI engineers Charles Frye and David Wang optimized FlashAttention-4 for large language model inference, focusing on decode-heavy workloads dominated by memory bandwidth-limited token generation. The team adjusted p…