Intel names D-Series GPU for AI inference with up to 480GB memory
Intel has officially named its AI-focused Crescent Island GPU as the D-Series, targeting inference workloads in server environments. The card features 32 Xe3P cores, 256 XMX engines, and supports up to 480GB of LPDDR5X memory—with a default 160GB configuration for Intel-branded cards. It delivers 1.3 PetaFLOPS of FP8 performance and includes a media engine for multimodal tasks like video…
Key points
- Intel’s D-Series GPU supports **480GB LPDDR5X memory** (160GB standard for Intel-branded cards)
- Designed for AI inference, with **1.3 PetaFLOPS FP8 performance** and no graphics processing
- Supports models up to **300B parameters** on one card, **trillion-parameter clusters** with 8x GPUs
The Xe3P architecture introduces a 16-deep systolic pipeline in XMX engines, supports FP4 to FP64 data formats, and lacks traditional graphics processing. Intel claims the card can run models with up to 300 billion parameters on a single unit, with an 8x cluster handling trillion-parameter models. Sampling is underway, but pricing, availability, and benchmark results remain undisclosed. The card runs on PCIe 5.0 x16 with 350W power draw, requiring no liquid cooling.
The story so far
2 episodes →- Intel names D-Series GPU for AI inference with up to 480GB memorythis story
Intel "Crescent Island" GPU Gets D-Series Name, Supports Up to 480 GB of LPDDR5X
techpowerup.com · 12 October 2026
Loading the full article…
This text was published by techpowerup.com. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
1source- Reddit discussionreddit.com
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Lambda integrates NVIDIA’s Open-Source Run:AI scheduler into its Kubernetes service · 1 src
- Apple launches HomeView smart display with Siri AI for $299 · 3 src
- Apple adds Siri AI and Wi-Fi 7 to fourth-gen Apple TV 4K for $299 · 2 src
- Microsoft expands Azure Maia system to multi-gigawatt deployments starting in 2027 · 4 src
- Nvidia’s GPU demand drives record revenue and new access options · 1 src
Comments
via GitHub Discussions