WSE
WSE（Wafer Scale Engine，晶圆级引擎）是一种将整块硅晶圆制造为单一完整芯片的处理器架构技术。与传统将晶圆切割成多个独立芯片再封装互联的方式不同，WSE通过保留晶圆完整性，大幅增加片上计算核心数量、内存容量及内部带宽，同时消除芯片间互联的通信延迟与功耗损耗，主要应用于大规模人工智能训练和高性能计算领域。
Core Facts
Cerebras Systems specializes in wafer-scale chips (WSE), making an entire silicon wafer into a single processor, with on-chip memory bandwidth far exceeding that of traditional GPU clusters.
Cerebras Systems focuses on wafer-scale chips (Wafer-Scale Engine), fabricating an entire silicon wafer into a single processor
Timeline (last 90 days)
Cerebras's WSE integrates hundreds of billions of transistors and hundreds of thousands of AI compute cores, significantly reducing the latency and energy consumption caused by cross-chip communication.
WSE的90万个核心通过片上2D mesh网络互联，核间通信带宽达到220 PB/s
All Facts (4)
Cerebras Systems specializes in wafer-scale chips (WSE), making an entire silicon wafer into a single processor, with on-chip memory bandwidth far exceeding that of traditional GPU clusters.
90%VerifiedCerebras Systems focuses on wafer-scale chips (Wafer-Scale Engine), fabricating an entire silicon wafer into a single processor
75%UnverifiedCerebras's WSE integrates hundreds of billions of transistors and hundreds of thousands of AI compute cores, significantly reducing the latency and energy consumption caused by cross-chip communication.
50%UnverifiedWSE的90万个核心通过片上2D mesh网络互联，核间通信带宽达到220 PB/s
50%