NVIDIA DSX POD

NVIDIA's modular Vera Rubin AI factory reference design, cutting on-site deployment from six months to 14 days.

NVIDIA · AI hardware and silicon · infrastructure

Published in Feed

Overview

NVIDIA unveiled the DSX POD at GTC Berlin on October 21, 2026, a reference design for complete Vera Rubin AI factories that partners assemble and validate in modular containers before delivery. NVIDIA says on-site deployment drops from six months to 14 days once site utilities are ready. The companion DSX AI Factory Platform links the grid, the facility and the compute inside as one system, with DSX Sim and SimReady assets to simulate the grid, substation and AI factory before construction, DSX Flex to coordinate demand with grid conditions and onsite generation, DSX MaxLPS and DSX Exchange to improve compute per watt and share signals, and a DSX Ready qualification for power equipment. Partners include GE Vernova, Hitachi Energy, Schneider Electric, Siemens Energy, Phaidra (about 10% more compute in the same power envelope, per Phaidra) and Emerald AI. A third-party recap reports delivery planned for early 2027.

This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.

Key features

  • Modular containers integrated and validated by partners before shipping
  • On-site deployment time cut from 6 months to 14 days, per NVIDIA
  • DSX Sim, DSX Flex, DSX MaxLPS, DSX Exchange and DSX Ready components in the DSX AI Factory Platform
  • Hitachi Energy battery storage qualified as DSX Ready; Schneider Electric validated Vera Rubin power and cooling reference architectures
  • Claimed up to 100% less cooling water and up to 30% less power for the same output, per NVIDIA
  • Delivery planned for early 2027 (per a third-party keynote recap, not NVIDIA's blog)

Use cases

  • Standing up sovereign AI factories in Europe
  • Operators needing faster time to first production tokens
  • Planning data centers jointly with grid constraints

Pricing

Unknown; one commentary estimate cited about $50 billion for a 1GW AI factory including land and power, which is not an NVIDIA price.

Pricing reflects the source record and may change. Confirm current terms with the provider.

Dates and sources

Released
21 Oct 2026
Record updated
27 Oct 2026
Record last checked
27 Oct 2026

Explore the Feed

Other entries filed under ai hardware and silicon; inclusion does not imply equivalent features.

  • Azure Maia 200 and Maia 300

    Microsoft says its Maia 200 AI accelerator is in production serving Copilot, and unveiled Maia 300 for 2027 deployment.

  • NVIDIA Vera Rubin NVL144

    NVIDIA's next-generation rack-scale AI system pairing Vera CPUs with Rubin GPUs, detailed with a shipping timeline pulled forward to compete with AMD's newly announced Helion rack.

  • AMD Helion

    AMD's open rack-scale blueprint for agentic AI infrastructure, with CPU-only configurations up to 32,768 EPYC cores and an integrated model-call token router.

Browse the Feed