Nvidia's Blackwell AI 'superchip' is the most powerful yet.
Nvidia has unveiled its most powerful "superchip" for training artificial intelligence models, called the Blackwell B200 graphics processing unit (GPU). The Blackwell chip has 208 billion transistors, compared to the 80 billion transistors of Nvidia's current-generation Hopper chips. Nvidia claims t...
| Publicado en: | New Scientist Vol. 261; no. 3483; p. 11 |
|---|---|
| Autor principal: | |
| Formato: | Journal Article |
| Publicado: |
New Scientist Ltd.
3/23/2024
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=176140526&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 176140526 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 02624079 NSI jtl: New Scientist issn: 02624079 maglogo: N pubinfo: dt: 3/23/2024 vid: 261 iid: 3483 pid: 39714 pub: New Scientist Ltd. artinfo: ui: 176140526 176140526 ppf: 11 ppct: 0 formats: tig: atl: Nvidia's Blackwell AI 'superchip' is the most powerful yet. aug: au: Hsu, Jeremy sug: ab: Nvidia has unveiled its most powerful "superchip" for training artificial intelligence models, called the Blackwell B200 graphics processing unit (GPU). The Blackwell chip has 208 billion transistors, compared to the 80 billion transistors of Nvidia's current-generation Hopper chips. Nvidia claims that the Blackwell chips can deliver a 30 times performance improvement while using 25 times less energy compared to the Hopper GPUs. However, the cost of the Blackwell GPUs is likely to be high, making them accessible only to select organizations and countries. pubtype: Periodical doctype: Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|