Apple is teaming with Nvidia to enhance the performance speed of artificial intelligence (AI) models.
As per multiple outlets, Apple recently revealed that it has been exploring different ways to speed up AI models by using Nvidia’s platform.
The goal is to enhance both the efficiency and speed of large language models (LLMs).
Apple used a technique called Recurrent Drafter (ReDrafter) which was introduced in a research paper earlier this year, along with Nvidia’s TensorRT-LLM framework to speed up the process of AI models.
Sources revealed that Apple researchers described their collaboration with Nvidia to enhance the performance of large language models (LLMs).
Earlier this year, Apple introduced the ReDrafter technique as an open-source tool, introducing a new way to speed up the decoding of data.
The researchers reveal that this method can make the process of generating tokens in large language models up to 3.5 times faster per generation step.
Although, Apple was able to improve the efficiency to some extent, they found that speed didn’t improve significantly.
To solve this issue, the researchers integrated the ReDrafter technique with Nvidia’s TensorRT-LLM interference acceleration framework to enhance the overall performance and speed.
