Japanese company is promising to cut prices for AI computing to provide the world’s most affordable GPU cloud

  • October 21, 2023
  • 2 min read

Third Wave wants to undercut hyperscalers in it’s Raytrek Cloud service

Japanese company Third Wave is launching a “GPU cloud” service for those who want to run generative AI applications – such as big language models (LLMs) or simulations as well as computer-aided design (CAD) on their own personal computers.

You don’t need the most powerful GPUs or PCs to make use of Third Wave’s cloud services, named Raytrek Cloud, with the company leasing indirectly the computing capabilities needed to tap the potential that is AI as it gets more popular.

The company revealed its GPU cloud during Dospara Plus Synapse 2023, as well as confirming that it is working on workstations and servers for AI. It will be in the form of a cloud service as well as the development and distribution of the equipment.

Utilizing low-cost AI workloads on the cloud

The company also announced the introduction of five brand new offerings which include the X4630 workstation, equipped with a 4th-generation Xeon Scalable processor, as well as NVIDIA’s RTX6000 Ada GPU.

This is in line with the company’s recent reorganization, which has resulted in its GALLERIA name for its gaming and consumer PCs, and Raytrek for its hardware designed for enterprise.

The introduction of Raytrek Cloud, meanwhile, is in light of the company’s realization that we’re set for another increase in the amount of data over the next years, with sixty times more data available today than it was ten years ago.

Director of Third Wave, executive officer in charge and vice president of Third Wave Akiya Ida added that data centers being used for cloud computing have led to problems with energy consumption, with the future AI systems expected to run on cloud as well as on the premises.

In Raytrek Cloud, businesses can connect to a remote computer via a desktop as a service (DaaS) offering jobs like the CAD process or modeling of buildings (BIM). There is also the possibility of using the GPU service to run AI applications, like as intelligent AI and simulations that require high-performance computing.

The company ultimately hopes to be the lowest-cost AI cloud service provider, with fees that are about a third to a quarter of a cent than other cloud providers according to Computer Monitor. The service is also pay-as-you-go, meaning you can use it on an hourly basis, or in smaller amounts.

About Author


Leave a Reply

Your email address will not be published. Required fields are marked *