Google Gemma team partners with Hugging Face to boost inference speed 5x

The post China is giving Manus to Tencent after blocking Meta’s attempted acquisition appeared on BitcoinEthereumNews.com. China is pushing Manus toward a new group of owners after telling Meta Platforms (NASDAQ: META) to reverse its $2 billion purchase of the AI agent startup. As per the Financial Times, Tencent Holdings (HKEX: 0700) will hold the maximum stake, but not more than 50 percent. It means that Manus will remain an independent entity and will continue its operations from Singapore instead of becoming a part of Tencent. The new deal would keep the same $2 billion valuation used when Meta bought Manus in December 2025. Former investors such as Tencent, ZhenFund, and HSG are taking part in the talks, along with Manus executives. Other investors could still join before an agreement is reached. China forces investors to rebuild Manus after Meta separates the business Beijing ordered Meta to reverse the purchase in April 2026, saying the deal broke local investment rules. Manus c

The US policy shift strengthens UAE tech ties, boosting AI and crypto sectors, but geopolitical risks could reverse these gains if trust falters. The post US eases export controls on advanced chips and drones for UAE in major policy shift appeared first on Crypto Briefing.

SK Hynix's NASDAQ debut highlights the growing importance of AI infrastructure, potentially reshaping global investment dynamics in tech. The post SK Hynix makes historic NASDAQ debut amid AI-driven demand appeared first on Crypto Briefing.

The post NVIDIA CUDA Kernel Fusion Boosts GPU Efficiency in AI Workloads appeared on BitcoinEthereumNews.com. Timothy Morano Jul 10, 2026 17:21 NVIDIA’s CUDA kernel fusion cuts memory traffic, kernel launch overhead, and speeds up AI and HPC tasks by up to 3x. Key for MoE and LLM training. NVIDIA (NASDAQ: NVDA) is doubling down on GPU efficiency with its latest advancements in CUDA kernel fusion, a technique that optimizes memory usage and minimizes kernel launch overhead. By combining multiple operations into a single kernel, NVIDIA claims speedups of up to 3x for certain workloads like Mixture-of-Experts (MoE) models and large language model (LLM) training. Kernel fusion works by addressing a longstanding bottleneck in GPU computing: the high memory bandwidth consumption caused by intermediate data transfers. In a typical GPU workload, intermediate results often travel through global memory between separate kernel launches, creating significant latency. Fusion elimi

The post NVIDIA Pushes Hardware-Aware LLM Co-Design for AI Efficiency appeared on BitcoinEthereumNews.com. Rongchai Wang Jul 10, 2026 17:13 NVIDIA highlights hardware-friendly AI model design to optimize LLM performance. Learn how co-design boosts throughput, latency, and cost-efficiency. NVIDIA has detailed its approach to hardware-aware large language model (LLM) design, a strategy that simultaneously optimizes AI model architectures and the hardware they run on. This co-design approach aims to maximize throughput, reduce latency, and lower costs for LLM deployments across data centers and edge devices. The blog post, published on July 10, 2026, provides practical guidelines for aligning AI models with modern GPU capabilities. Co-design focuses on balancing three core metrics for AI performance: accuracy, throughput, and interactivity. For instance, NVIDIA stresses that small design choices, such as aligning model dimensions with GPU tile sizes or choosing wider mod

The post Singapore: Resilient GDP and strong NODX – DBS appeared on BitcoinEthereumNews.com. DBS economists Radhika Rao and Mo Ji forecast Singapore’s advance 2Q26 Gross Domestic Product (GDP) growth at 5.8% year-on-year and 1.5% quarter-on-quarter seasonally adjusted, slightly below 1Q26 but still resilient. They cite strong manufacturing and wholesale trade on AI-related electronics demand, robust modern services and construction, and expect non-oil domestic exports to post a fourth consecutive month of double-digit growth despite a slowdown from May. AI demand underpins growth outlook “We expect Singapore’s advance GDP growth estimate for 2Q26 to register 5.8% yoy, 1.5% qoq sa, remaining resilient compared with 6.0% yoy, 1.0% qoq sa in 1Q26.” “Manufacturing accelerated, while wholesale trade performed well despite some moderation, driven by robust global demand for artificial intelligence (AI)-related electronics.” “Modern services remained resilient, supported by continued momentum

The post OpenAI sold AI to Pentagon-blacklisted Chinese firms, raising alarms appeared on BitcoinEthereumNews.com. OpenAI has reportedly supplied AI technology to Chinese companies on the Pentagon’s military-linked blacklist, adding fresh scrutiny to U.S. controls over advanced artificial intelligence exports. Summary OpenAI and Google reportedly provided AI access to Chinese firms on the Pentagon’s Section 1260H blacklist. The reported access has renewed debate over U.S. AI export controls and cloud-based model distribution. The development comes as OpenAI expands GPT-5.6 Sol, Terra, and Luna across ChatGPT, Codex, and its API. According to the reported findings, OpenAI and Google provided access to their AI models to Chinese companies included on the U.S. Department of Defense’s Section 1260H list, which identifies entities the Pentagon believes are tied to China’s military-industrial complex. While inclusion on the list does not automatically prohibit commercial dealings or trigger

OpenAI has reportedly supplied AI technology to Chinese companies on the Pentagon’s military-linked blacklist, adding fresh scrutiny to U.S. controls over advanced artificial intelligence exports. According to the reported findings, OpenAI and Google provided access to their AI models to…