Sakana AI, a Tokyo-based startup, has announced a groundbreaking system set to revolutionize artificial intelligence (AI) training and deployment. Unveiled on Thursday, this innovative solution promises to enhance the speed of AI development by up to 100 times. Co-founded in 2023 by two former Google engineers, Sakana AI is poised to make significant strides in the AI industry with this new technology.
The core of Sakana AI's advancement lies in its ability to automate the creation of code that controls Nvidia's graphic processing units (GPUs), which are renowned for their superior performance in AI applications. By leveraging Nvidia's CUDA technology—a programming model and API for general-purpose computing on GPUs—the system significantly accelerates both the training and inference phases of AI. This leap in efficiency addresses the increasing demand for faster and more efficient AI development.
Nvidia's GPUs are pivotal in AI due to their high-performance capabilities and parallel processing. Sakana AI's system capitalizes on these strengths, offering a tenfold to hundredfold improvement in speed. Such advancements are expected to benefit a range of industries reliant on AI, including healthcare, finance, and transportation, by facilitating quicker and more efficient AI processes.
The name "Sakana," meaning "fish" in Japanese, reflects the company's innovative approach to navigating the complexities of AI technology. The new system represents a major response to the growing need for streamlined AI development, addressing both current and future demands within the industry.
Leave a Reply