Intel "Crescent Island" GPU Gets D-Series Name, Supports Up to 480 GB of LPDDR5X
The newly named D-Series GPU has up to 480GB of LPDDR5X memory, 1.5TB/s of memory bandwidth
Proof2 community threads
Key points
- Intel rates it for up to 1.3 PetaFLOPS of FP8 performance, with support for formats ranging from FP4 to FP64.
- No cost estimates currently, I hope they launch a generic blower style variant at a later date for prosumer workstations.
- Intel GPU D-Series Specifications Intel GPU D270
- Memory (Max) 480GB LPDDR5X (ODMs), 160GB LPDDR5X (Intel)
Sources (2)
- [1]Intel "Crescent Island" GPU Gets D-Series Name, Supports Up to 480 GB of LPDDR5Xr/LocalLLaMA (top, daily) · Oct 12, 10:42 PM
The newly named D-Series GPU has up to 480GB of LPDDR5X memory, 1.5TB/s of memory bandwidth
Intel rates it for up to 1.3 PetaFLOPS of FP8 performance, with support for formats ranging from FP4 to FP64.
- [2]Intel D270 GPU AI Accelerator–Up to 480 GB LPDDR5XHacker News (AI, 100+ points) · Oct 12, 09:35 PM
Extractive summary: sentences quoted from the sources.
Before this
- Oct 12, 2026openbmb/MiniCPM-V-4.7-1B
- Oct 12, 2026openbmb/MiniCPM-V-4.7-35B-A3B
- Oct 12, 2026U-Space: Uncovering When and Why Uncertainty Arises in Language Models [R][D]
- Oct 12, 2026Qwen3.8-Flash-Next on six 3090s with vLLM: 7 users at 262k, 6.5k t/s prefill, 57 t/s decode solo / 206 t/s aggregate with all 7 generating. Fixes for the upstream included.
- Oct 12, 2026Inside the AI Economy: What Stripe's Data Reveals — Nilofer Rajpurkar, Stripe
- Oct 11, 2026Agents That Own Their Inference — Du'an Lightfoot & Khaja Omer, Akamai Technologies