2 publications · 2 reports

Intel unveils Intel GPU D-Series AI accelerators with 1.3 PFLOPS FP8 performance

The Next Interrupt briefing

AI-generated · Updated · Oct 13, 2026 · How we work

Intel announced its Crescent Island AI accelerators will be officially named the Intel GPU D-Series, disclosing a theoretical peak performance of 1.3 PFLOPS of FP8 compute and a memory bandwidth of 1.5 TB/s at OCP 2026. [1][2]

What changed

The chips are built on the Xe3P graphics architecture in an air-cooled PCIe form factor with a 350W TDP and up to 480GB of LPDDR5X memory. [1]

What remains uncertain

Intel stated that customer sampling of GPU D-Series products is underway, but did not provide details for a broad launch. [1]

Across the publications

PublicationOrder2 publications

Chips & hardware illustration · ServeTheHome

[2] ServeTheHomePublisher coverage

Oct 12, 2:25 PM CT

Intel’s Crescent Island AI Server GPU Gets a Name and Specs: Meet the Intel GPU D-Series

Intel today has unveiled its official performance specifications and branding for its upcoming "Crescent Island" data center AI inference cards. Going under the name Intel GPU D-Series, the D270 will offer 1.3 PLFOPs of dense FP8 and 1.5TB/second of memory bandwidth

Excerpt · ServeTheHome

Read at ServeTheHome

Tom’s Hardware · Intel

[1] Tom’s HardwarePublisher coverage

Oct 12, 2:48 PM CT

Intel's Crescent Island AI accelerator officially powers new GPU D-Series products

Intel's Crescent Island AI inference accelerator will come to market as the GPU D-Series family, promising 1.3 PFLOPS of FP8 compute alongside up to 480 GB of LPDDR5X delivering 1.5 TB/s of memory bandwidth.

Excerpt · Tom’s Hardware

Read at Tom’s Hardware