Layer-Streaming Telemetry Harness
Benchmark massive MoE LLMs under strict 0GB VRAM limits
什么是 Layer-Streaming Telemetry Harness？
An open-source MIT diagnostic tool to track peak system RAM, active VRAM allocation, and data-transfer layer execution when offloading extreme 284B parameter models onto commodity hardware footprints. Built for hardware-agnostic architecture audits.
截图
?
还没有评论，来抢沙发吧！
X 上关于 Layer-Streaming Telemetry Harness 的真实讨论
去 X 发帖