Posts in CPU （46）

Alibaba Unveils New AI Chip "Zhenwu V900," Eyes 10-Trillion-Parameter Qwen Models

Meta Begins Testing In-House AI Inference Chip "Arke," With Next-Gen "Astrid" Nearing Completion

MediaTek Unveils Dimensity 9600 Pro: First 2nm Flagship SoC Cuts Multi-Core Power by 61%, Adds LPDDR6 and UFS 5.0

Fujitsu Unveils Domestic 2nm CPU FUJITSU-MONAKA, Doubling AI Inference Throughput Ahead of November Server Launch

* Image is for illustration purposes.

SAMSUNG and Mistral AI Partner on AI for Chip Design and Manufacturing

OpenAI and Samsung Deepen Partnership on Next-Generation AI Chips, Adding Memory and Stargate Cooperation

RIKEN and IBM Simulate a 12,635-Atom Protein on Quantum Computers, Reaching the Gordon Bell Prize Finals

ASML and TSMC Launch Industry Initiative to Move to 12-Inch Photomasks, Pilot Line Targeted for 2031

How TSMC Changed in a Decade: From a 16nm Contract Fab to a US$2 Trillion AI Foundation

AMD Unveils 96-Core Threadripper Halo Station at IFA 2026 Keynote for Trillion-Parameter Local AI

IBM Unveils 2nm Dual-Architecture Processor: One Core Runs Both Arm and IBM Z Instructions

Intel Eyes a Memory Comeback 40 Years On, With XBM and ZAM Aimed at the HBM Cost Structure

SAMSUNG Raises Advanced Foundry Prices by Up to 15 Percent as AI Demand Fills Its 4nm Lines

Qualcomm Teases Two Next-Gen Snapdragon Flagships, Reveal Set for September 22

Zhaoxin KH-50000 Takes Two Top Prizes at China Internet Innovation Contest

Zhaoxin ZX86 Spreads Across China's Classrooms: 1 Million Units Shipped, 34,000 Won in Hainan

NVIDIA Deploys Its Own Vera CPU to Speed Up Design of Next-Gen CPUs and GPUs

AMD Unveils 6th Gen EPYC Venice CPUs: Up to 256 Zen 6 Cores, First HPC Chip on TSMC 2nm

Ryzen 7 7700X3D Arrives in Japan at 63,980 Yen, Priced Above the Faster 7800X3D

Zhaoxin Shows Off Its ZX86 AI Computing Stack in Shanghai, From 96-Core KH-50000 Servers to Edge Devices

Zhaoxin Shows Off Its ZX86 AI Lineup, From the 96-Core KH-50000 to the Edge

Zhaoxin Partners with Dongfang Suanxin on ZX86 CPU and DF1000 3D AI Chip to Build a Homegrown Compute Base

NVIDIA Details Vera CPU Design Philosophy: 1.8x x86 Per-Core Performance for Agents, Perplexity Eyes Deployment

* Image is for illustration purposes.

Meta Reuses Old DDR4 Memory in DDR5-Only AI Servers, Cutting Server Count by Up to 25% with Its Custom Vistara Chip