Microsoft is expanding its Phi Silica AI model to Nvidia RTX 30-series and newer GPUs with at least 6GB of VRAM in an early developer preview, bringing local AI capabilities to more Windows users beyond Copilot+ PCs, though GPU-based execution lacks some NPU-exclusive features like prompt compression and speculative decoding.
Qualcomm unveils the Snapdragon Reality Elite chip at Augmented World Expo, delivering 48 TOPS of on-device AI performance, 4.4K-per-eye visuals at 90fps, and 20% better battery life to power next-gen AR headsets, including the newly launched XREAL Project Aura glasses running Google's Android XR platform.
Atom Computing secures $100 million in Series C funding led by Third Point Ventures, cementing its position as one of only nine U.S. government-backed quantum computing startups and signaling surging investor confidence in the quantum computing sector.
Snap launches SPECS, a $2,195 pair of AR glasses powered by dual Snapdragon chips and contextual AI, offering a 51-degree field-of-view display and up to 20 hours of battery life, now available for pre-order ahead of a fall release in the US, UK, and France.
CoreWeave unveils 'Valvey' and 'Racky,' two groundbreaking infrastructure innovations designed to eliminate AI data center bottlenecks by revolutionizing coolant control and rack management for Nvidia's powerful 72-GPU Vera Rubin NVL72 system.
A full GPT transformer implemented entirely in RTL Verilog is now running on a Xilinx Virtex-5 FPGA, blazing through 69,200 tokens per second at 80 MHz after engineers overcame two critical synthesis bugs that silently zeroed ROM arrays and folded live registers, causing the board to hang despite passing simulation.
Smart glasses sales explode 167% year-over-year in Q1 2026, with 2.25 million units shipped in a single quarter as Meta dominates with a 69.2% market share, while Google, Snap, and Chinese rivals intensify competition in a race toward mainstream adoption projected to hit 13.6 million units by year's end.
NVIDIA's GB300 NVL72 dominates the industry's first agentic AI benchmark, AgentPerf, delivering a staggering 20x more AI agents per megawatt than its predecessor, the HGX H200, as top inference providers race to deploy the powerful Blackwell architecture for real-world production applications.
LMCache, an open-source KV cache management layer for LLM inference, surpasses 9,100 GitHub stars as it transforms temporary GPU cache into persistent, reusable knowledge—cutting time-to-first-token and boosting throughput across major hardware vendors like NVIDIA and AMD.
Oracle shares plunge up to 11% as the tech giant reveals plans to raise $40 billion in debt and equity financing amid a staggering $23.7 billion in negative free cash flow, spooking investors despite a strong earnings beat with revenue surging 21% to $19.18 billion and capital expenditures skyrocketing 162% …