Facebook parent Meta unveils chip for AI programs
Facebook parent company Meta has unveiled its first custom silicon chip for processing AI programs called the Meta Training and Inference Accelerator, or MTIA. It also talked about a new AI-optimised data centre design and the second phase of 16,000 GPU supercomputers for AI research.
According to the company, the new efforts in the direction of AI, along with the additional projects still underway, will enable it to develop larger, more sophisticated AI models and then deploy them efficiently at scale.
“We’re even reimagining how we code by deploying CodeCompose, a generative AI-based coding assistant we developed to make our developers more productive throughout the software development lifecycle,” the company said in a statement.
With the new AI infrastructure, Meta wants to create a scalable foundation to power opportunities in areas like generative AI and the metaverse.
What is MTIA
MTIA is an in-house, custom accelerator chip family targeting inference workloads. It is claimed to provide greater compute power and efficiency than CPUs, and it is customised for the company’s internal workloads.
According to Meta, both MTIA chips and GPUs will help it deliver better performance, decreased latency and greater efficiency for each workload.
MTIA is supported by the company’s AI data centre that will not only support the current products but also enable future generations of AI hardware for both training and inference.
The data centre will be an AI-optimised design, support liquid-cooled AI hardware and a high-performance AI network connecting thousands of AI chips together for data centre-scale AI training clusters.
Microsoft, Samsung also building AI chips
Microsoft and Samsung are also said to be working on AI chips. Recently, a report said that Microsoft is working with AMD on AI chips to counter Nvidia.
A separate report claimed that Samsung and Naver plan to unveil AI chips in the second half of this year to compete with Nvidia’s graphics processing unit (GPU), which powers AI platforms. Nvidia currently dominates the market for AI-capable chips.
FacebookTwitterLinkedin
end of article
Facebook parent company Meta has unveiled its first custom silicon chip for processing AI programs called the Meta Training and Inference Accelerator, or MTIA. It also talked about a new AI-optimised data centre design and the second phase of 16,000 GPU supercomputers for AI research.
According to the company, the new efforts in the direction of AI, along with the additional projects still underway, will enable it to develop larger, more sophisticated AI models and then deploy them efficiently at scale.
“We’re even reimagining how we code by deploying CodeCompose, a generative AI-based coding assistant we developed to make our developers more productive throughout the software development lifecycle,” the company said in a statement.
With the new AI infrastructure, Meta wants to create a scalable foundation to power opportunities in areas like generative AI and the metaverse.
What is MTIA
MTIA is an in-house, custom accelerator chip family targeting inference workloads. It is claimed to provide greater compute power and efficiency than CPUs, and it is customised for the company’s internal workloads.
According to Meta, both MTIA chips and GPUs will help it deliver better performance, decreased latency and greater efficiency for each workload.
MTIA is supported by the company’s AI data centre that will not only support the current products but also enable future generations of AI hardware for both training and inference.
The data centre will be an AI-optimised design, support liquid-cooled AI hardware and a high-performance AI network connecting thousands of AI chips together for data centre-scale AI training clusters.
Microsoft, Samsung also building AI chips
Microsoft and Samsung are also said to be working on AI chips. Recently, a report said that Microsoft is working with AMD on AI chips to counter Nvidia.
A separate report claimed that Samsung and Naver plan to unveil AI chips in the second half of this year to compete with Nvidia’s graphics processing unit (GPU), which powers AI platforms. Nvidia currently dominates the market for AI-capable chips.
FacebookTwitterLinkedin
end of article
Denial of responsibility! Techno Blender is an automatic aggregator of the all world’s media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, all materials to their authors. If you are the owner of the content and do not want us to publish your materials, please contact us by email – [email protected]. The content will be deleted within 24 hours.