DIRS.INFO top alerts submit data jurnalism about API
login
Top related persons:
Dado Ruvic - 1 Alexis Black Bjorlin - 1 Yee Jiun - 1 Black Bjorlin - 1 Katie Paul - 1 Stephen Nellis - 1 Kenneth Li - 1 Lisa Shumaker - 1
Top related locs:
Top related orgs:
REUTERS - 1 Meta - 1 Broadcom - 1 Intel - 1 Nvidia - 1 Thomson - 1

Search resuls for: "Katie Paul Stephen Nellis"


2 mentions found


☆ Exclusive: Meta executive leading AI chip efforts to leave position - sources
  + stars: | 2023-09-27 | by ( Katie Paul Stephen Nellis | Katie Paul | Stephen Nellis | ) www.reuters.com   time to read: +2 min
A smartphone with a displayed Meta logo is placed on a computer motherboard in this illustration taken February 23, 2023. She is leaving her position at the end of the month but not immediately leaving the company, one of the sources said. Meta on Wednesday introduced the company's first generative AI products for consumers, including a chatbot that can generate both text responses and photo-realistic images. Some of the company's initial efforts proved slower than using existing chips, leading the company to scrap some of its AI chips. Meta is working on a newer chip that will span all types of AI work.
Persons: Dado Ruvic, Alexis Black Bjorlin, Yee Jiun, Black Bjorlin, Katie Paul, Stephen Nellis, Kenneth Li, Lisa Shumaker Organizations: REUTERS, Meta, Broadcom, Intel, Nvidia, Reuters, Thomson
☆ Meta announces AI training and inference chip project
  + stars: | 2023-05-18 | by ( Katie Paul Stephen Nellis | Katie Paul | Stephen Nellis | ) www.reuters.com   time to read: +3 min
Reuters previously reported that the company was not planning to deploy its first in-house AI chip widely and was already working on a successor. The blog posts portrayed the first MTIA chip as a learning opportunity. As part of that, the company scrapped plans for a large-scale rollout of an in-house inference chip and started work on a more ambitious chip capable of performing both training and inference, according to the Reuters reporting. Meta acknowledged in its blog posts that its first MTIA chip stumbled with high-complexity AI models, although it said the chip handled low- and medium-complexity models more efficiently than competitor chips. The MTIA chip also used only 25 watts of power - a fraction of what market-leading chips from suppliers such as Nvidia Corp (NVDA.O) consume - and used an open-source chip architecture called RISC-V, Meta said.
0 next >>
Total: 2