Meta Introduces AI Software Solutions to make Transitioning between AMD and Nvidia CPUs easier
Meta has released free software tools for AI applications to easily switch between different chips
Meta Platforms Inc, the parent company of Facebook, said Monday that it has released a new suite of free software tools for AI applications that can make it easier for developers to switch between different core chips.
The new open-source AI platform Meta is built on an open-source machine learning framework called PyTorch and can help code run up to 12 times faster on Nvidia Corp’s flagship A100 chip or four times faster. Faster on Advanced Micro Devices Inc. MI250 chip said.
Meta said in a blog post that the flexibility that the software can provide is just as important as the increased speed.
The software has become a major battleground for chip makers looking to build an ecosystem for developers to use their chips. Nvidia’s CUDA platform has been by far the most popular in AI work.
However, once developers port their code to Nvidia chips, it’s harder to run on GPUs, or GPUs, than Nvidia competitors like AMD. Meta said the software is designed to easily switch between chips without being locked.
“Unified GPU backend support gives deep learning developers more options for hardware vendors with minimal migration costs,” Meta said in its blog post.
Nvidia and AMD did not immediately respond to requests for comment.
The Meta software is built for AI work called inference, which is when machine learning algorithms that have been pre-trained on large amounts of data are called to make quick judgments, such as determining whether an image is a cat or a bird.
“This is a cross-platform software effort,” said David Kanter, founder of MLCommons, an independent group that measures speed. “It is a testament to the importance of software, particularly for implementing neural networks in inference machine learning.” Artificial intelligence.
Kanter added that this new Meta AI platform would be “good for customer choice.”
Read More News Updates Visit our Site InsightFello.