NVDA 162.85 ▼2.01%GOOGL 295.77 ▼0.98%MSFT 523.88 ▲0.17%AMD 534.06 ▼8.75%INTC 59.66 ▼2.52%TSMC 398.73 ▼1.67%AMZN 266.66 ▼0.29%META 641.69 ▲2.90%AAPL 356.27 ▲0.39%PLTR 167.99 ▼2.09%
Markets at last close

Intel · Chips

Intel previews Crescent Island GPU and courts Korean partners

·1 min read

Intel unveiled a mock-up of its Crescent Island GPU at its AI Summit in Seoul on Nov. 12, expanding its infrastructure lineup beyond CPUs. The chip is under development, with a launch targeted for next year, and is designed for high-capacity, low-power AI computing without expensive high bandwidth memory.

According to Intel, Crescent Island uses LPDDR5X memory and supports up to 480 gigabytes (GB), offering capacity up to 5 times larger than Nvidia’s RTX Pro 6000. Intel puts power usage at 350 watts (W), compared with 600W for the Nvidia product. Intel executive Anil Nanduri emphasized scalable inference, saying easily expandable GPUs provide flexibility to run models from small to ultra-large.

Intel also pitched Korean companies on collaboration spanning chip design, advanced packaging and foundry services, including joint development with customers that already have their own designs. It announced the K-AI Full Stack Alliance alongside 23 domestic companies, linking hardware development with industry-specific services. Participants include AI developer Upstage, NPU companies FuriosaAI and DeepX, and service developer Hancom.

Originally reported by businesskorea.co.krRead the source →
Related coverage
All Intel news →