$Intel (INTC.US)$ Gaudi 3 has 40% faster time-to-train large...
Gaudi 3 has 40% faster time-to-train large language models versus NVIDIA’s H100 AI chip and is 50% faster inferencing on LLMs versus the NVIDIA H100, Intel said.
The Gaudi 3 will launch with Dell, Hewlett Packard Enterprise, Lenovo and Supermicro as OEM partners.
Disclaimer: Community is offered by Moomoo Technologies Inc. and is for educational purposes only.
Read more
Comment
Sign in to post a comment
73176442 : Do you really believe Intel?! Don’t forget about 10nm+![[undefined]](https://static.moomoo.com/nnq/emoji/static/image/default/default-black.png?imageMogr2/thumbnail/36x36 "undefined")
70141151 : Funding has responded to the latest chips...
NMK19 : Dont believe.
peter kim344 : no credit
Lucci Loon : Sound fishy
Money Thrill : Sorry, I do not believe this, And it is not a GPU?Nvidia dominates the GPU markets for data center and machine learning applications – and was quick to leverage its generative AI capabilities. And the next generation is nano with quantum Light technology with ASML , who works together with NVIDIA. For INTEL, it is too late