NVIDIA released NeMo AutoModel, an open-source library that accelerates fine-tuning of Mixture-of-Experts (MoE) transformer models by 3.4-3.7x in training throughput and reduces GPU memory usage by 29…
Polar Signals storage engineer wrote a CUDA kernel to decompress FSST-encoded strings on GPUs, but initial performance was worse than CPU decompression. Using the company's GPU profiler with PC sampli…
Yann LeCun, Meta's chief AI scientist, argued at the United Nations Open Source Week that open-source AI is the only viable path to global AI sovereignty, cultural diversity, and long-term safety, war…
Congressman Sam Liccardo introduced the SKILL Act on Monday at West Valley College, creating tax credits for companies that invest in community college workforce programs to address job displacement f…
NVIDIA has released a GenAI LLM Certification Lab that guides developers through building a production-ready fine-tuning and optimization pipeline. The lab covers data preparation, LoRA fine-tuning wi…
Google's TPU v5p and upcoming Ironwood TPU7x offer higher raw matrix throughput and simpler interconnects than NVIDIA H100 GPUs for large-scale AI training, but GPUs remain superior for inference with…
Seltz, a nine-person startup, raised $12.5 million in seed funding to build a search infrastructure purpose-built for AI agents, replacing legacy APIs like Google and Bing. The company wrote its own c…
AtkinsRéalis, a Canadian engineering firm, is seeking US regulatory approval for its CANDU nuclear reactor technology to power AI data centers. The company's nuclear revenue has grown to 25% of total …
Amazon's Prime Day sale offers discounts on gaming laptops, including models from Acer, ASUS, Alienware, and MSI with RTX 50-series GPUs and latest processors. Deals feature up to 21% off, with prices…
NVIDIA's shift to an 800V power distribution architecture for AI data centers is redirecting semiconductor investment toward power delivery components like GaN and SiC devices. The transition, expecte…
Gartner predicts that by 2030, GPU-specialized neocloud providers will capture 20% of the $267 billion AI cloud market, representing roughly $53 billion shifting away from AWS, Azure, and Google Cloud…
Amazon Prime Day GPU deals are now live, with top picks from reviewer Sebastian Kozlowski including the PNY RTX 5080, ASUS RTX 5060 Ti, and Gigabyte RX 9070 XT at discounts of 5-8%.…
NVIDIA open-sourced Grove, a Kubernetes API for managing multi-component AI inference stacks, at KubeCon Europe 2026. Grove introduces custom resources for gang scheduling, topology-aware placement, a…
The European Commission has selected Italian startup Domyn to lead the Frontier AI Grand Challenge, a project to build a 400-billion-parameter language model aimed at achieving European technological …
Alienware's 16 Aurora gaming laptop with an RTX 5070 GPU dropped to $2,099.99 on Amazon, its lowest price in months, as part of Prime Day deals. The 18% discount applies to a model featuring an Intel …
UC San Diego researchers developed DFlash, a speculative decoding method that uses a lightweight block diffusion model to draft entire token blocks in parallel, achieving up to 6.08x speedup on Qwen3-…
MSI's Vector A18 HX 18-inch gaming laptop with an RTX 5080 and Ryzen 9 9955HX dropped to a record-low price of $2,332.79 during Amazon Prime Day, a 14% discount. The deal includes 32GB DDR5 RAM, a 1TB…
Biostar announced the EdgeComp MS-NAT5000, a compact edge AI computing system integrating the NVIDIA Jetson Thor T5000 module for generative AI, visual intelligence, and real-time sensor processing in…
Photoroom reduced inference latency of its convolutional segmentation backbone by 22% on A100 GPUs by switching to PyTorch's channels-last memory format. The change required only four lines of code an…
Yann LeCun's AMI Labs raised $1.03B at a $3.5B valuation to build world models using JEPA, signaling a shift from LLMs to physics-aware AI. NVIDIA's Cosmos platform, trained on 20 million hours of rea…