At CES 2024 earlier this week, Nvidia pushed for a new category of PCs, called AI PCs, which are souped up with advanced components and computing power to accelerate AI tasks on-device.
It released the GeForce RTX 40 Super family of graphics processors to supercharge the latest games and form the core of AI-powered PCs. This product lineup draws on the company’s advancing GPU technology, particularly in the area of AI-enhanced gaming and applications.
They are the GeForce RTX 4080 Super, GeForce RTX 4070 Ti Super and GeForce RTX 4070 Super. All offer enhanced performance over their predecessors.
GeForce RTX Super cards support over 500 RTX games and applications and will have users prepared for a new wave of generative AI apps coming to PC. It is the products’ ability to unlock the full potential of AI on Windows PCs which marks a milestone in PC industry.
The new GeForce RTX Super GPUs have the computational power to accelerate AI processes on-device. Specialised AI Tensor Cores in the GPUs deliver up to 836 AI TOPS (trillions operations per second) for AI in gaming, creating and everyday productivity tasks. The rich software stack built on top of RTX GPUs further accelerates AI.
Nvidia TensorRT is software for high-performance deep learning inference, which includes a deep learning inference optimiser and runtime that delivers low latency and high throughput for inference applications.
TensorRT-LLM for Windows is an open-source library that accelerates inference performance for the latest large language models. In AI workloads, Nvidia says the GeForce RTX 4080 Super generates video over 1.5x faster and images over 1.7x faster than the RTX 3080 Ti.
In gaming, the AI-powered DLSS (deep learning super sampling) – a neural graphics technology – will keep providing greater in-game immersion, years after its initial launch. Meanwhile, GenAI applications like Adobe Photoshop can take advantage of Tensor Cores to speed productivity and keep creative workflows moving.
The GeForce RTX 4080 Super, priced at US$999, is set to hit the market on Janyary 31. Launching on Jan 24, the GeForce RTX 4070 Ti Super is priced at US$799. It will receive a significant upgrade with 16GB of memory, up from the 12GB on its predecessor.
The GeForce RTX 4070 Super, set to be released on January 17 at US$599, is likely to appeal to a broader range of buyers due to its lower price. The cards are available at Nvidia.com and retailers.
The trio of new Super cards are now being used in PCs made by Asus, Lenovo, Acer, Samsung, HP and Razer.
Meanwhile at the JPMorgan Chase & Co healthcare conference in San Francisco earlier this week, Nvidia disclosed that it is collaborating with biotech firm Amgen to build AI models to fuel among other things, new drug discoveries.
Amgen will use the AI model-building platform, named Freyja, at the office of its subsidiary called deCODE Genetics in Reykjavik, Iceland. The system aims to analyse extensive human datasets for drug targets and biomarker discovery, facilitating diagnostics for disease progression and regression.
Nvidia also announced at the conference that its BioNeMo service generative AI platform is now in beta, featuring fresh foundational models which can be put to use by biotech and pharmaceutical companies to optimise research.