Array
(
[content] =>
[params] => Array
(
[0] => /forum/whats-new/latest-activity?before_id=93570
)
[addOns] => Array
(
[DL6/MLTP] => 13
[Hampel/TimeZoneDebug] => 1000070
[SV/ChangePostDate] => 2010200
[SemiWiki/Newsletter] => 1000010
[SemiWiki/WPMenu] => 1000010
[SemiWiki/XPressExtend] => 1000010
[ThemeHouse/XLink] => 1000970
[ThemeHouse/XPress] => 1010570
[XF] => 2021770
[XFI] => 1050270
)
[wordpress] => /var/www/html
)
You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please,
join our community today !
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser .
K
They have a facility called MemoryX units, which are connected via their proprietary inter-WSE interconnect called SwarmX. This article...
Sep 4, 2025
K
Iirc their largest stakeholder (70+%?) is in the Middle East.
I like their approach, I think it's a clever way to work with what we...
Sep 4, 2025
B
Sounds like a lot of below cost product could be hitting the market.
Lots of lunch gonna get eaten.
How long you think the Govts both...
Sep 4, 2025
B
Tough sledding in that middle sector.
Its why all TSMC results should be in a separate sector to the rest , they skew the headlines.
Sep 4, 2025
F
My SSD has 4 trillion transistors flash chips
Sep 4, 2025
M
Iirc their largest stakeholder (70+%?) is in the Middle East.
I like their approach, I think it's a clever way to work with what we...
Sep 4, 2025
F
SK Hynix did 5 layers for VLSI 2024: https://www.businesskorea.co.kr/news/articleView.html?idxno=219741
Sep 4, 2025
In theory their tech should give them the best perf/watt -- which should let them scale down cost per token in a large environment...
Sep 4, 2025
I don't know how to answer this. Cerebras offers their own cloud services for their systems, and the pricing is publicly advertised...
Sep 4, 2025
Amdahl, Amdahl, Amdahl was indeed the absolute star, and hype champion of the era, and we know how he ended. You could indeed have a...
Sep 4, 2025
K
In theory their tech should give them the best perf/watt -- which should let them scale down cost per token in a large environment...
Sep 4, 2025
I agree - they have found a way to build wafer scale chips that have killed many companies before them and harness the WSE-3 / CSE-3 on...
Sep 4, 2025
Of course, it depends on the application, but performance matters in the usual metrics: latency (time to first token), and throughput...
Sep 4, 2025
Does outright performance really matter here though?
AI is already a highly risky investment for any major firm, and if you're spending...
Sep 4, 2025
It's a game changer for AI. Nothing touches it, performance-wise, especially for inference. There are numerous challenges for a...
Sep 4, 2025
In theory their tech should give them the best perf/watt -- which should let them scale down cost per token in a large environment...