Intel has shared further technical details about Crescent Island, a dedicated AI inference accelerator based on its Xe3P ...

Testing on SemiAnalysis’ InferenceX benchmark suite — presumably this is an unofficial test — shows OpenAI’s Jalapeño-based ...

Intel, in a detailed presentation, detailed the package design of its next-generation Xeon 7 "Diamond Rapids" server ...

Qualcomm reveals its next-gen Oryon CPU, crossing the 5GHz threshold for the first time in mobile to power advanced agentic AI workloads.

Intel unveiled key package design and architecture details of its next-generation Xeon 7 Diamond Rapids processors, featuring a massive architecture overhaul with up to 16 CPU tiles and 1.28GB cache.

The chip is designed specifically for AI inference — the stage where trained models generate answers, rather than the ...

Sandisk Corporation is upgraded to Strong Buy, driven by its high-bandwidth flash targeting AI inference memory bottlenecks.

Joo Young-pyo, SK hynix's executive vice president for system architecture, struck a cautious tone on the timing of the CXL ...

Qualcomm is previewing three components of its “next-generation Snapdragon flagship platform” (chip), starting with the Oryon ...

Getting 3D-stacked chiplets to chat with each other is hard, but so is making them connect to other chips in the system.

Intel's long-awaited graphics card update is arriving for data centers and workstations with the introduction of "Crescent ...

Cerebras CS-4, unveiled at SUPERNOVA 2026, doubles AI inference speed to 4,400 tokens per second on the same 5-nanometer ...