LLM inference in C/C++

A fast multi-producer, multi-consumer lock-free concurrent queue for C++11

纯c++的全平台llm加速库，支持python调用，chatglm-6B级模型单卡可达10000+token / s，支持glm, llama, moss基座，手机端流畅运行

C++ IPC Library: A high-performance inter-process communication using shared memory on Linux/Windows.

NVIDIA Data Center GPU Manager (DCGM) is a project for gathering telemetry and measuring the health of NVIDIA GPUs

Microsoft Collective Communication Library

MSCCL++: A GPU-driven communication stack for scalable AI applications

Automatic virtualization of (general) accelerators.