“It was trained on a bit more than 3K GPU, so 10 megawatt of power, that's, we think, three to four times less than the Chinese competitors”

The MAD Podcast: How AI Gets Built - with Matt Turck · Mistral Is SO Back: Inside Le Chonk, the West’s Biggest Open Model (Pierre Stock, VP Science, Mistral)

Play from 19:05
19:05

Audio streams from the publisher’s own feed. Verbatim does not host or copy it.

What was described

trained the model on a bit more than 3K GPUs consuming about 10 MW of power, which is three to four times less than Chinese competitors

≈3,000 GPUs; 10 MW; 3-4× less power than Chinese rivals

Said nearby on the same subject

“The Chinese staff work four times harder than the Vietnam staff.”
♫
Said on The Ecomcrew Ecommerce Podcast
Hard number Chinese staff work four times harder than Vietnamese staff Share this moment
“Even at three agents, what we found was three agents were not aligning on a simple goal more than 16% of the time.”
VP
Vijoy Pandey Cisco
DataFramed ·
Hard number 16% misalignment Share this moment
“We don't really do any that are less than 90 square meters internally now… downsizers… want a two bedroom apartment… just that little bit more generous”
♫
Said on The Deal with Adam Di Marco
Why it worked 90 m² Share this moment
“You’ll get the base level of 15%, but then you’ll get the 33% discount on the 15%. So it comes down to about 10% on capital gains, which is far less than what the government's proposing of the minimum of 30%”
♫
Said on Business Essentials
Hard number about 10% CGT Share this moment

Moments like this, every Monday

The strongest new moments on your topics, in the speakers' own words. Free for three topics.

No card. No password. One-click unsubscribe.

Track a competitor?

Pro watches up to 25 names or phrases and emails you the morning after they are said.

See Pro, $29 a month

Verbatim quotes a short extract and links to the episode; the episode belongs to its publisher. Remove a show.