AI Chips news and analysis on Web Pulse: 6906 curated articles tracking the latest AI Chips developments, tools, and research, updated continuously from vetted sources.
CoolIT is deploying modular coldplate liquid-cooling loops, refined over six generations of fanless designs, to handle near-total heat capture in AI server racks exceeding 250 kW. The company says rising processor therma…
Qualcomm will fully unveil its Snapdragon 8 Elite Gen 6 smartphone chip at the Snapdragon Summit on September 22 in Maui, with the processor able to run a 30 billion parameter mixture-of-experts AI model entirely on-devi…
Qualcomm unveiled the Snapdragon 8 Elite Gen 6 and Snapdragon 8 Elite Extreme Gen 6 at its annual Snapdragon Summit, claiming the chips' Prime cores reach a 5GHz clock speed, up from 4.6GHz on the previous-generation Sna…
Qualcomm unveiled the Snapdragon 8 Elite Extreme Gen 6 and Snapdragon 8 Elite Gen 6 mobile chips at Snapdragon Summit 2026 in Maui, Hawaii, both built on a 2nm process node. The 8 Elite Extreme Gen 6 delivers 13% improve…
Qualcomm announced the Snapdragon 8 Elite Gen 6 and Snapdragon 8 Elite Extreme Gen 6, a pair of flagship 2nm chips built on an Oryon CPU with two 5GHz Prime cores and six 4GHz Performance cores, for the next generation o…
Qualcomm announced two flagship smartphone processors at its annual Snapdragon Summit, the Snapdragon 8 Elite Gen 6 and Snapdragon 8 Elite Extreme Gen 6, both built to run AI agents and models on-device. The Extreme vers…
Qualcomm launched two new flagship smartphone chips, the Snapdragon 8 Elite Extreme Gen 6 and Snapdragon 8 Elite Gen 6, at its annual Snapdragon Summit in Maui, Hawaii on Tuesday, with the company saying both are designe…
Federal agencies are failing to deploy AI at scale because they conflate AI training with inference, even though as much as 90% of enterprise and most government AI workloads are inference tasks that can run on existing …
Amazon's custom silicon program has advanced from the Nitro System through Graviton to the AI chips Inferentia, Trainium2, Trainium3, and a roadmap to Trainium4, with each generation driven by customer workloads rather t…
A user submitted a formal feature suggestion to the Google AI Product Team proposing a "Compute Intensity Selector" for Gemini that would add an Eco Mode forcing lightweight models such as Gemma 2B or a Flash-Lite varian…
Creative Strategies published an expert interview with Billy on the power constraints facing AI data center buildouts, examining behind-the-meter generation, natural gas supply, and contract structuring. The interview fo…
NVIDIA released Topograph, an open source toolkit that discovers cluster network topology from cloud APIs or on-premises fabric systems and publishes it as Kubernetes node labels, Slurm topology configuration, or Slinky …
AliCloud unveiled an "Agentic Cloud" stack at the 2026 Apsara Conference, including AgentCore, Agent Sandbox, and a new CPFS storage layer, to support models such as Qwen3.8 Max and KIMI K3 that exceed 2T parameters. Ali…
STMicroelectronics is demonstrating four Physical AI applications at Embedded World North America, including a stable robotic hand and a virtual humanoid walking simulator built on NVIDIA Isaac Sim and Isaac Lab, using S…
NVIDIA launched its DSX Ready qualification program for data center power and cooling hardware, covering battery energy storage systems (BESS) and cooling distribution units (CDUs) that NVIDIA does not itself supply. LG …
Chinese chipmaker Hygon Information Technology launched its Hygon 1000 series CPUs on Tuesday, expanding beyond cloud computing into embedded processors for robotics and industrial edge devices. The four-core, eight-thre…
VAST Data launched DataEnclave, software that creates a hardware-isolated secure runtime with cryptographic attestation so enterprises can run AI models against sensitive data while model builders keep proprietary weight…
Astrus, a Khosla-backed startup, has raised new funding to automate analog chip design using reinforcement learning, targeting one of the biggest bottlenecks in chip design. The post With new funding, Astrus prepares to …
Ornn Data reported that the cheapest qualifying open-weight model on the Artificial Analysis Intelligence Index completes a task at roughly one fifth the cost of a comparable closed model, and that self-hosting on rented…
HPE and NVIDIA are pushing liquid cooling as a strategic imperative for AI data centers, citing rack densities of 70 kW to 100 kW+ required for next-generation NVIDIA Blackwell GPUs that air cooling cannot handle. The co…