Intel launches D-Series AI GPUs with up to 1.3 PFLOPS
Intel has officially named its Crescent Island chip the D-Series GPU and revealed new technical specs. The server-focused AI accelerator now claims up to 1.3 PFLOPS of FP8 compute at 350W, with memory options up to 480GB using LPDDR5X. It's built purely for AI inference—no graphics features—and fits in standard air-cooled PCIe slots. Customer sampling has started but pricing and real-world benchmarks aren’t available yet.
- Rebranded Crescent Island as Intel D-Series
- 1.3 PFLOPS FP8 performance at 350W
- Supports up to 480GB LPDDR5X memory
- Designed for server AI inference, not graphics
- Aims for models up to 300 billion parameters
Sources covering this
In this story
More in Chips
Microsoft unveils Maia 300 AI chip, plans to scale deployments
Microsoft just unveiled its Maia 300 AI chip, the third version of its in-house silicon, and said it will shift most of its AI services…
AMD unveils Helion rack for agentic AI with up to 32,768 cores
AMD just announced the Helion rack, a new open design for building AI server systems.
Nvidia and Cisco announce ultra-dense 400Gb/s USFP optical modules
Nvidia and Cisco unveiled a new optical module called USFP, designed to deliver 400Gb/s per lane and up to 12.8Tb/s of network capacity.
Nvidia shares Vera Rubin AI rack designs with OCP
Nvidia is opening up its latest Vera Rubin rack and compute tray designs for AI data centers, contributing them to the Open Compute…