Ascend Python Packages

uc-manager

Persist and reuse KV Cache to speedup your LLM.

vllm-ascend

Community maintained hardware plugin for vLLM on Ascend

nndeploy

一款简单易用和高性能的AI部署框架 | An Easy-to-Use and High-Performance AI Deployment Framework

torch-rechub

A Lighting Pytorch Framework for Recommendation Models, Easy-to-use and Easy-to-extend.

llm-cal

LLM inference hardware calculator — architecture-aware (MLA/NSA/MoE), engine-aware (vLLM/SGLang), honest-labeled. Reads real safetensors bytes; supports 53 GPUs (NVIDIA / AMD / Huawei Ascend / 沐曦 / 昆仑芯 / 壁仞 / 寒武纪 / 海光).

ascend-nputop

An interactive Ascend-NPU process viewer

cmq-test

vLLM Ascend backend plugin