News

Big NVIDIA Generative AI Push Led by GPUs, 'AI Workbench' Dev Tooling

• By David Ramel
• 01/08/2024

NVIDIA, already seen by many as the leader in processors and other PC hardware/software for AI, unveiled a big new generative AI push that includes new GPUs, developer tools and more.

"Generative AI is the single most significant platform transition in computing history and will transform every industry, including gaming," said CEO/founder Jensen Huang during the CES 2024 event. "With over 100 million RTX AI PCs and workstations, NVIDIA is a massive installed base for developers and gamers to enjoy the magic of generative AI."

The Jan. 8 announcement post is chock full of new offerings, including:

• GeForce RTX SUPER Desktop GPUs: NVIDIA announced the release of GeForce RTX SUPER desktop GPUs, aimed at enhancing generative AI performance.
• AI Laptops: New AI laptops from top manufacturers were introduced, equipped with NVIDIA RTX-accelerated AI software and tools.
• Generative AI Software and Tools: NVIDIA unveiled new software and tools for developers and consumers to leverage generative AI on RTX PCs and workstations.
• NVIDIA AI Workbench: This new toolkit, set for beta release later this month, is designed to aid AI developers in creating, testing and customizing AI models and large language models (LLMs).
• RTX AI PCs and Workstations: NVIDIA highlighted the capabilities of their RTX GPUs in running a broad range of applications, particularly for generative AI.
• TensorRT-LLM: An open source library for accelerating and optimizing inference performance of large language models, now supporting more pre-optimized models for PCs.
• Generative AI-Powered Applications and Services: NVIDIA and its partners are releasing new AI-powered applications and services for PCs, including tools for game remastering and digital avatars, as well as enhanced performance for popular AI models.

Of special interest to developers is the AI Workbench, set to debut soon in beta. It helps developers access popular repositories like Hugging Face, GitHub and NVIDIA NGC, the company said, along with a simplified UI to help those devs more easily reproduce, collaborate on and migrate projects.

"Running generative AI locally on a PC is critical for privacy, latency and cost-sensitive applications," the company said. "It requires a large installed base of AI-ready systems, as well as the right developer tools to tune and optimize AI models for the PC platform."