Technology / WORLD

Intel announces data-center GPU D-series, says customer sampling is under way

Intel announced its data-center GPU D-series and new open rack designs at the OCP Global Summit.

Reading sizeSources ↓Share story ↗
Conceptual illustration of technology, processors and connected information.Illustration
Symbolic illustration

The Intel-branded D-series PCIe card has 160 GB of LPDDR5X memory; the company says partner ODM card designs offer memory options up to 480 GB. The announcement lists 1.5 TB/s memory bandwidth and up to 1.3 petaflops of FP8 performance in a 350-watt power envelope.

Intel said customer sampling is under way; the announcement does not say general sales or volume shipments have begun. These performance figures are manufacturer-stated peak specifications.

Data

Table

Scope of the announced design

Intel announced its data-center GPU D-series and new open rack designs at the OCP Global Summit.

Memory bandwidth12000000000000
Methodology

Records GPU memory and peak-performance specifications in Intel’s product announcement; these are not independent benchmark results. Intel’s source gives memory bandwidth as 1.5 TB/s. The table uses the decimal SI prefix (1 TB = 10¹² bytes) and 8 bits per byte: 1.5 × 10¹² × 8 = 12 × 10¹² bit/s. This is a unit conversion only.

Updated:

Structured coverage

Subjects and places

Subject proposals

Geographic references

Only source-provided geographic references are shown. Opening an external map is optional. Optional story maps use OpenStreetMap. Opening a map sends your IP address and this site’s address to the map provider; location data remains readable as text without loading the map.

  • United States · Place role: Coverage · Place type: Country or area · Location precision: Regional
    Evidence
    • Explicit preserved editorial country/region; no city or national centroid inferredwww.intc.com ↗
Taxonomy version: IPTC Media Topics 2026-Q2

Continue reading

Technology

Intel announces GPU D-series accelerators for AI

At the OCP Global Summit, Intel introduced GPU D-series accelerators aimed at data-center inference and agentic AI workloads. The company said the product was previously known by the codename Crescent Island and customer sampling is underway.

Intel1 min read
Data included