Intel unveils 480GB GPU D-series for AI workloads at Open Compute Project summit
Intel announced the new GPU D-series with 480GB of memory at the recent Open Compute Project Global Summit.
New memory capacity record set
The graphics processor features up to 480GB of LPDDR5X memory. This configuration surpasses the Nvidia Blackwell Ultra B300, which provides 288GB of HBM3e memory.
Technology choice prioritizes efficiency
Intel selected LPDDR5X instead of GDDR7 to focus on power-conscious design. The company aims to support demanding AI inference tasks rather than gaming applications.
Large model support capabilities
A single card can handle models with up to 300 billion parameters. An eight-GPU system is capable of accommodating models exceeding one trillion parameters.
Reported by one outlet
Only one outlet has published this. Nothing here has been checked against a second report, so read it as that outlet's account and follow the link below for the original.
Reported by
1 independent outlet. Headline as published. Links open the original report.