gpus
AMD Unveils ROCm 6.1 Software for Radeon GPUs, Enhancing AI Development
AMD releases ROCm 6.1 software, extending AI capabilities to Radeon desktop GPUs, enabling scalable AI solutions and broader support for AI frameworks.
NVIDIA H100 GPUs and TensorRT-LLM Achieve Breakthrough Performance for Mixtral 8x7B
NVIDIA's H100 Tensor Core GPUs and TensorRT-LLM software demonstrate record-breaking performance for the Mixtral 8x7B model, leveraging FP8 precision.
Luminary Cloud Accelerates Engineering Simulations with NVIDIA GPUs
Luminary Cloud leverages NVIDIA GPUs to speed up engineering simulations, addressing industry challenges and enhancing productivity.
AMD Instinct MI300X Accelerators Boost Performance for Large Language Models
AMD's MI300X accelerators, with high memory bandwidth and capacity, enhance the performance and efficiency of large language models.
AMD Unveils ROCm 6.2.3 Enhancing AI Performance on Radeon GPUs
AMD releases ROCm 6.2.3, boosting AI capabilities for Radeon GPUs with enhanced support for Llama 3, Stable Diffusion, and Triton framework, improving AI development efficiency.
NVIDIA GPUs Revolutionize Quantum Dynamics Simulations
Researchers utilize NVIDIA GPUs to enhance quantum dynamics simulations, overcoming computational challenges and enabling advancements in quantum computing and material science.
NVIDIA's AI and RTX GPUs Revolutionize Reality Capture
NVIDIA leverages AI and RTX GPUs to enhance reality capture technologies like NeRFs and Gaussian splatting, improving 3D modeling and visualization processes.
Llama 3.1 405B Achieves 1.5x Throughput Boost with NVIDIA H200 GPUs and NVLink
NVIDIA's latest advancements in parallelism techniques enhance Llama 3.1 405B throughput by 1.5x, using NVIDIA H200 Tensor Core GPUs and NVLink Switch, improving AI inference performance.
Microsoft Develops Secret AI Chips to Reduce Development Costs
Microsoft has been developing its own AI chips since 2019 to reduce reliance on Nvidia’s GPUs due to rising costs. The project, called “Athena,” is already being tested by Microsoft’s machine-learning staff and OpenAI developers.
Voltage Park's $1 Billion Cloud Infrastructure Targets ML Compute Shortage
Voltage Park, launching on October 29, 2023, introduced a $1 billion cloud infrastructure housing around 24,000 NVIDIA H100 GPUs, aimed at redressing the ML compute resource crunch. A subsidiary of the Navigation Fund, Voltage Park aspires to democratize ML compute access, thereby catalyzing AI innovation across the spectrum.
Elon Musk Moves Forward with AI Plans for Twitter
Elon Musk’s recent purchase of nearly 10,000 graphics processing units (GPUs) indicates his commitment to an AI project at Twitter. The project is reportedly in its early stages and uses a large language model. Musk has previously expressed concerns about AI and signed an open letter to halt its development.
Alibaba Unveils Its First Home-Grown AI Chip
Chinese e-commerce giant Alibaba unveiled its first artificial intelligence inference chip on Wednesday, a move which could further invigorate its already rip-roaring cloud computing business.