Nvidia and Google Cloud have released G2 Virtual Machines (VMs) with Nvidia L4 GPUs, a universal accelerator for efficient video, AI, and graphics.
This joint offering aims to enable organizations to adopt the latest AI models while delivering top performance and lowering costs.
Google has unveiled G2, a new class of cloud GPUs, which is powered by the Nvidia L4 Tensor Core GPU, and is purpose-built for large inference AI workloads.
- Nvidia L4 is currently available on Google Cloud through Google’s G2 virtual machines.
- L4 is said to deliver 120x more AI-powered video performance than CPUs while improving energy efficiency by 99%.
- Platform users can access real-time insights through L4 to personalize content, improve search relevance, detect unsuitable content, and implement use-case-specific solutions through an easy-to-use interface.
- The new offering is also equipped with capabilities for video decoding, transcoding, and streaming.
