---
title: 安装与验证
description: "1. 配置环境变量。"
url: https://www.hikunpeng.com/document/detail/zh/kunpengai/ecosystemEnable/flashattention/kunpengflashattention_04_0005.html
sourcePath: /source/zh/kunpengai/ecosystemEnable/flashattention/kunpengflashattention_04_0005.html
indexId: d416493ea8977bb96ea8c8eadb63d7956a7418165d10e3090359f931c270041286
---
# 安装与验证

1. 配置环境变量。
  1 2 3 4 5 6 7 export CUDA_HOME=/usr/local/cuda-13.0 export PATH=$CUDA_HOME/bin:$PATH export LD_LIBRARY_PATH=$CUDA_HOME/lib64:${LD_LIBRARY_PATH:-} export FLASH_ATTENTION_FORCE_BUILD=TRUE export FLASH_ATTN_CUDA_ARCHS=80 export MAX_JOBS=8 export NVCC_THREADS=1

2. 使用pip命令安装FlashAttention。
  1 python3 -m pip install flash-attn==2.8.3 --no-build-isolation

3. 验证FlashAttention是否安装成功。
  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 python3 - <<'PY' import torch print("torch_version=" + torch.__version__) print("torch_cuda=" + str(torch.version.cuda)) print("cuda_built=" + str(torch.backends.cuda.is_built())) print("cuda_available=" + str(torch.cuda.is_available())) assert torch.backends.cuda.is_built() assert torch.cuda.is_available() import flash_attn print("flash_attn_version=" + flash_attn.__version__) assert flash_attn.__version__ == "2.8.3" PY

  验证结果预期输出如下信息则安装成功。

  1 2 3 4 torch_cuda=13.0 cuda_built=True cuda_available=True flash_attn_version=2.8.3