home.social

#graceblackwell — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #graceblackwell, aggregated by home.social.

  1. #AMD is launching #Helios, its first #rackscaleAI system, to compete with #Nvidia’s #GraceBlackwell and #VeraRubin systems. Microsoft, Meta, OpenAI, Oracle, and Tata Consultancy Services are among the companies adopting Helios, which combines AMD’s #GPUs, #CPUs, #networking, #software. Helios aims to offer the lowest #costpertoken and significant benefits over Nvidia’s systems, potentially increasing AMD’s market share in the data centre GPU market. cnbc.com/2026/07/20/amd-helios #tech #media #news

  2. #AMD is launching #Helios, its first #rackscaleAI system, to compete with #Nvidia’s #GraceBlackwell and #VeraRubin systems. Microsoft, Meta, OpenAI, Oracle, and Tata Consultancy Services are among the companies adopting Helios, which combines AMD’s #GPUs, #CPUs, #networking, #software. Helios aims to offer the lowest #costpertoken and significant benefits over Nvidia’s systems, potentially increasing AMD’s market share in the data centre GPU market. cnbc.com/2026/07/20/amd-helios #tech #media #news

  3. #AMD is launching #Helios, its first #rackscaleAI system, to compete with #Nvidia’s #GraceBlackwell and #VeraRubin systems. Microsoft, Meta, OpenAI, Oracle, and Tata Consultancy Services are among the companies adopting Helios, which combines AMD’s #GPUs, #CPUs, #networking, #software. Helios aims to offer the lowest #costpertoken and significant benefits over Nvidia’s systems, potentially increasing AMD’s market share in the data centre GPU market. cnbc.com/2026/07/20/amd-helios #tech #media #news

  4. #AMD is launching #Helios, its first #rackscaleAI system, to compete with #Nvidia’s #GraceBlackwell and #VeraRubin systems. Microsoft, Meta, OpenAI, Oracle, and Tata Consultancy Services are among the companies adopting Helios, which combines AMD’s #GPUs, #CPUs, #networking, #software. Helios aims to offer the lowest #costpertoken and significant benefits over Nvidia’s systems, potentially increasing AMD’s market share in the data centre GPU market. cnbc.com/2026/07/20/amd-helios #tech #media #news

  5. #AMD is launching #Helios, its first #rackscaleAI system, to compete with #Nvidia’s #GraceBlackwell and #VeraRubin systems. Microsoft, Meta, OpenAI, Oracle, and Tata Consultancy Services are among the companies adopting Helios, which combines AMD’s #GPUs, #CPUs, #networking, #software. Helios aims to offer the lowest #costpertoken and significant benefits over Nvidia’s systems, potentially increasing AMD’s market share in the data centre GPU market. cnbc.com/2026/07/20/amd-helios #tech #media #news

  6. The funding reflects how integral #AI has become to #NationalSecurity. The #tech has helped #military & #spy agencies sift through massive amounts of #intelligence & is particularly valuable for tasks like finding overlooked communications intercepts.
    The $9B is intended in part to boost the availability of #infrastructure that can support #Nvidia’s #GraceBlackwell superchip, which requires #DataCenters that can supply enormous amounts of electrical #energy & specialized liquid #cooling systems.

  7. The funding reflects how integral #AI has become to #NationalSecurity. The #tech has helped #military & #spy agencies sift through massive amounts of #intelligence & is particularly valuable for tasks like finding overlooked communications intercepts.
    The $9B is intended in part to boost the availability of #infrastructure that can support #Nvidia’s #GraceBlackwell superchip, which requires #DataCenters that can supply enormous amounts of electrical #energy & specialized liquid #cooling systems.

  8. The funding reflects how integral #AI has become to #NationalSecurity. The #tech has helped #military & #spy agencies sift through massive amounts of #intelligence & is particularly valuable for tasks like finding overlooked communications intercepts.
    The $9B is intended in part to boost the availability of #infrastructure that can support #Nvidia’s #GraceBlackwell superchip, which requires #DataCenters that can supply enormous amounts of electrical #energy & specialized liquid #cooling systems.

  9. The funding reflects how integral #AI has become to #NationalSecurity. The #tech has helped #military & #spy agencies sift through massive amounts of #intelligence & is particularly valuable for tasks like finding overlooked communications intercepts.
    The $9B is intended in part to boost the availability of #infrastructure that can support #Nvidia’s #GraceBlackwell superchip, which requires #DataCenters that can supply enormous amounts of electrical #energy & specialized liquid #cooling systems.

  10. The funding reflects how integral #AI has become to #NationalSecurity. The #tech has helped #military & #spy agencies sift through massive amounts of #intelligence & is particularly valuable for tasks like finding overlooked communications intercepts.
    The $9B is intended in part to boost the availability of #infrastructure that can support #Nvidia’s #GraceBlackwell superchip, which requires #DataCenters that can supply enormous amounts of electrical #energy & specialized liquid #cooling systems.

  11. Dell's new Pro Max workstation, powered by the GB10 chip, brings on‑device AI to the edge—supporting massive 70‑billion‑parameter models without a cloud. Paired with NVIDIA’s Grace Blackwell architecture, it promises developers full freedom to train and run AI locally. Discover how this could reshape edge AI workflows. #DellProMax #GB10 #OnDeviceAI #GraceBlackwell

    🔗 aidailypost.com/news/dell-laun

  12. #Nvidia’s #VeraRubin Platform Obsoletes Current AI Iron Six Months Ahead Of Launch
    Nvidia’s top brass says Vera-Rubin #NVL72 #rackscale system, with 72 GPU sockets and 36 CPU sockets as well as NVSwitch fabric linking the, delivers a 10X reduction in #inference cost per token for MoE #AI models and 4X reduction in number of GPUs (but maybe not a 4X reduction in cost, be careful) to train those models compared to the prior generation of #GraceBlackwell NVL72 systems.
    nextplatform.com/2026/01/05/nv

  13. #Nvidia’s #VeraRubin Platform Obsoletes Current AI Iron Six Months Ahead Of Launch
    Nvidia’s top brass says Vera-Rubin #NVL72 #rackscale system, with 72 GPU sockets and 36 CPU sockets as well as NVSwitch fabric linking the, delivers a 10X reduction in #inference cost per token for MoE #AI models and 4X reduction in number of GPUs (but maybe not a 4X reduction in cost, be careful) to train those models compared to the prior generation of #GraceBlackwell NVL72 systems.
    nextplatform.com/2026/01/05/nv

  14. ’s Platform Obsoletes Current AI Iron Six Months Ahead Of Launch
    Nvidia’s top brass says Vera-Rubin system, with 72 GPU sockets and 36 CPU sockets as well as NVSwitch fabric linking the, delivers a 10X reduction in cost per token for MoE models and 4X reduction in number of GPUs (but maybe not a 4X reduction in cost, be careful) to train those models compared to the prior generation of NVL72 systems.
    nextplatform.com/2026/01/05/nv

  15. #Nvidia’s #VeraRubin Platform Obsoletes Current AI Iron Six Months Ahead Of Launch
    Nvidia’s top brass says Vera-Rubin #NVL72 #rackscale system, with 72 GPU sockets and 36 CPU sockets as well as NVSwitch fabric linking the, delivers a 10X reduction in #inference cost per token for MoE #AI models and 4X reduction in number of GPUs (but maybe not a 4X reduction in cost, be careful) to train those models compared to the prior generation of #GraceBlackwell NVL72 systems.
    nextplatform.com/2026/01/05/nv

  16. #Nvidia’s #VeraRubin Platform Obsoletes Current AI Iron Six Months Ahead Of Launch
    Nvidia’s top brass says Vera-Rubin #NVL72 #rackscale system, with 72 GPU sockets and 36 CPU sockets as well as NVSwitch fabric linking the, delivers a 10X reduction in #inference cost per token for MoE #AI models and 4X reduction in number of GPUs (but maybe not a 4X reduction in cost, be careful) to train those models compared to the prior generation of #GraceBlackwell NVL72 systems.
    nextplatform.com/2026/01/05/nv