Bullish
Intel's GPU D-Series AI accelerator unveiled, targeting 1.3 PFLOPS performance
Intel has officially introduced its GPU D-Series AI accelerator, built on the Xe3P architecture, offering a theoretical peak performance of 1.3 PFLOPS of FP8 compute and 1.5 TB/s memory bandwidth. This move signifies Intel's strategic entry into the AI inference market, aiming to support large-scale AI models and applications.
Key points
- Intel's GPU D-Series AI accelerator unveiled, targeting 1.3 PFLOPS performance
- Intel to phase out SGX after Xeon Diamond Rapids in favor of TDX
- Intel begins implementing Microsoft's Driver Quality Initiative for Windows 11
Sources
- Intel's Crescent Island AI accelerator officially powers new GPU D-Series productsTom's Hardware · October 12, 2026
- Intel to phase out SGX after Xeon Diamond Rapids in favor of TDXTom's Hardware · October 12, 2026
- Intel kicks off Windows 11's big driver overhaul - get ready for fewer bugsTechRadar · October 10, 2026
AI-generated from public news sources. Not financial advice.
