Nscale Signs $3.5 Billion Compute Deal With Figure
Nscale has signed a multiyear partnership to provide Figure with an initial $3.5 billion in computing resources, the companies announced in a press release. The agreement could grow beyond $6 billion and cover as many as 100,000 NVIDIA GPUs.
The first GPUs will use NVIDIA's Vera Rubin platform and are scheduled for deployment in Barstow, Texas, starting in the second half of 2027. The capacity will support training and deployment of Figure's Helix models and humanoid robots.
Nscale will become Figure's preferred compute provider and a shareholder. It is also making an investment of an undisclosed amount in Figure, while the companies will explore using humanoid robots in Nscale's supply chain.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Silicon Brief, Robotics Brief or Daily AI Brief.
Also, consider following us on social media:
More from Data Centers
Oct 21 Boeing Launches Dataliner AI Maintenance Platform With Microsoft Oct 21 SpaceX Q3 revenue doubles to $14.58 billion as AI compute rental brings in $7.6 billion Oct 21 SAP Q3 revenue passes €10 billion as cloud sales rise 26% Oct 21 Lam Research Posts Record $8.22 Billion Revenue as AI Chip Demand Rises Oct 21 Texas Instruments Q3 revenue rises 28% to $6.06 billionMore from Robotics
Oct 21 YYForce Plans Workforce OS Robotics Platform Using NVIDIA Tools Oct 21 GM Uses AI to Target Two Year Vehicle Development Oct 21 NVIDIA Expands Hyperion Autonomous Vehicle Platform in Europe Oct 21 NVIDIA Expands Isaac With Agile ONE SN Humanoid and Robot Learning Tools Oct 21 Agile Robots Plans Up to 2,000 Humanoids With NVIDIA in 2027Robotics Brief
Weekly coverage of AI-driven robotics advances in industrial automation, autonomous vehicles, and robotic systems.
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
Anthropic and Nscale Near Second $45 Billion Compute Deal
NVIDIA Partners Expand European AI Clouds and Agent Controls
Agile Robots Plans Up to 2,000 Humanoids With NVIDIA in 2027
General Compute Signs Cerebras Inference Deal
NVIDIA and Germany Plan Industrial AI Initiative
Daily AI Brief: the AI news that matters, in your inbox.