NVIDIA's new H200 Tensor Core GPUs and TensorRT LLM updates tackle soaring AI demands today

NVIDIA's new H200 Tensor Core GPUs and TensorRT LLM updates tackle soaring AI demands today

The trend is clear - generative AI model sizes are ballooning. Blackwell won't be here anytime soon, but Hopper GPUs with new memory and TensorRT advances will save the day. #genAI #generativeAI #ai #nvidia #hopper #h100 #h200 ##DataCenter #LLM
NVIDIA’s new 200-billion transistor Blackwell B200 GPU will tackle XXXL-sized generative AI models

NVIDIA’s new 200-billion transistor Blackwell B200 GPU will tackle XXXL-sized generative AI models

With a focus on accelerated computing, breakthrough data processing, and tackling trillion-parameter generative AI models, Blackwell is huge, both physically and in its promise for another way of change. #nvidia #blackwellgpu #gtc2024 #b200
AMD flexes HPC and AI superiority at Intel and NVIDIA with Genoa-X, Bergamo, and the Instinct MI300X GPU

AMD flexes HPC and AI superiority at Intel and NVIDIA with Genoa-X, Bergamo, and the Instinct MI300X GPU

AMD showcased an expanded 4th-gen EPYC CPU and Instinct MI300 lineup for dedicated technical, cloud native, and generative AI computing during their Data Centre and AI Technology Premiere Event in San Francisco.
AMD shows off 146-billion transistor Instinct MI300, the world's first integrated CPU and GPU data center chip (Updated)

AMD shows off 146-billion transistor Instinct MI300, the world's first integrated CPU and GPU data center chip (Updated)

ICYMI: To bring on exascale AI compute performance to the next level, the industry will need a breakthrough architecture. And AMD just reassured that it's coming this year.
NVIDIA adds liquid-cooled GPUs to lower power draw and cost in data center servers

NVIDIA adds liquid-cooled GPUs to lower power draw and cost in data center servers

For the first time, NVIDIA is launching liquid-cooled GPUs to target space-constrained data center servers and to lower power consumption. The NVIDIA A100 PCIe GPU will be the first of them.
NVIDIA’s 80-billion transistor H100 GPU and new Hopper Architecture will drive the world’s AI Infrastructure

NVIDIA’s 80-billion transistor H100 GPU and new Hopper Architecture will drive the world’s AI Infrastructure

6 breakthrough areas define the new Hopper Architecture and the H100 super chip, so find out what they are and the whole convoy of H100-based solutions.
Intel shadows NVIDIA; unveils Infrastructure Processing Unit (IPU) to drive up data center efficiency

Intel shadows NVIDIA; unveils Infrastructure Processing Unit (IPU) to drive up data center efficiency

Some 8 months later, Intel too has a solution for improving data center efficiency and is the first to be built in collaboration with hyerscale cloud partners.
NVIDIA's pro-grade GPUs adopt Ampere architecture for massive performance leaps

NVIDIA's pro-grade GPUs adopt Ampere architecture for massive performance leaps

Get ready for NVIDIA's upgraded professional GPU solutions to flood the desktop and laptop market, including options for remote visual processing needs.
Intel launches their 10nm 3rd Gen Xeon scalable 'Ice Lake' processors

Intel launches their 10nm 3rd Gen Xeon scalable 'Ice Lake' processors

With up to 40 cores per processor, Intel calls the new processor 'optimised' for AI, HPC, networking, security, and cloud workloads.
AMD’s new 3rd-gen EPYC processors looks set to continue its dominion over performance

AMD’s new 3rd-gen EPYC processors looks set to continue its dominion over performance

AMD's new third-gen server processors and Zen 3 architecture, EPYC 7003 series, look set to continue helming the price to performance crown against Intel.