安装xanylabeling1 方法一2 方法二背景 win11 RTX5080SAM3下载地址https://hf-mirror.com/vietanhdev/segment-anything-3-onnx-models/tree/main1 方法一git clone https://github.com/CVHub520/X-AnyLabeling.git cd X-AnyLabeling python-m venv venv venv\Scripts\activate pip install torch torchvision--index-url https://download.pytorch.org/whl/cu128 python-cimport torch; print(torch.__version__, torch.cuda.is_available(), torch.cuda.get_device_name(0))pip install-r requirements.txt python anylabeling/app.py2 方法二把下面内容保存为 install_xanylabeling.bat放在英文路径下如 D:\ai\双击运行。echooff chcp65001nul setlocal enabledelayedexpansion echoecho X-AnyLabeling 一键安装(RTX5080/CUDA12.8)echoecho.REM----------检查 Python----------python--versionnul21iferrorlevel1(echo[错误]未检测到 Python请先安装 Python3.10并加入 PATH pause exit/b1)for/ftokens2%%vin(python --version 2^^1)dosetPYVER%%v echo[信息]检测到 Python%PYVER%REM----------检查 Git----------git--versionnul21iferrorlevel1(echo[错误]未检测到 Git请先安装 GitforWindows pause exit/b1)REM----------安装目录----------setINSTALL_DIR%~dp0X-AnyLabelingifexist%INSTALL_DIR%(echo[信息]目录已存在:%INSTALL_DIR%set/p REUSE是否删除并重新克隆?(y/n):if/i!REUSE!y(rmdir/s/q%INSTALL_DIR%)else(echo[信息]复用现有目录 goto SKIP_CLONE))echo.echo[1/6]克隆 X-AnyLabeling 仓库...git clone https://github.com/CVHub520/X-AnyLabeling.git%INSTALL_DIR%iferrorlevel1(echo[错误]克隆失败请检查网络 pause exit/b1):SKIP_CLONE cd/d%INSTALL_DIR%REM----------创建虚拟环境----------echo.echo[2/6]创建虚拟环境...ifnotexistvenv(python-m venv venv)call venv\Scripts\activate.bat REM----------升级 pip----------echo.echo[3/6]升级 pip/setuptools/wheel...python-m pip install--upgrade pip setuptools wheel-i https://pypi.tuna.tsinghua.edu.cn/simple REM----------安装 PyTorch(CUDA12.8)----------echo.echo[4/6]安装 PyTorch(cu128,支持 RTX5080sm_120)...pip install torch torchvision--index-url https://download.pytorch.org/whl/cu128iferrorlevel1(echo[错误]PyTorch 安装失败 pause exit/b1)REM----------安装其余依赖----------echo.echo[5/6]安装项目依赖...REM 先固定 torch避免 requirements 把 torch 降级 pip install-r requirements.txt-i https://pypi.tuna.tsinghua.edu.cn/simpleiferrorlevel1(echo[警告]部分依赖安装失败尝试继续...)REM----------验证 GPU----------echo.echo[6/6]验证 GPU 环境...python-cimport torch; print(torch:, torch.__version__); print(cuda available:, torch.cuda.is_available()); print(device:, torch.cuda.get_device_name(0) if torch.cuda.is_available() else N/A)echo.echoecho 安装完成 echo 启动命令:echo cd/d%INSTALL_DIR%echo venv\Scripts\activate.bat echo python anylabeling\app.py echopause配套启动脚本 run_xanylabeling.bat再保存一个启动脚本双击即可运行echooff chcp65001nul cd/d%~dp0X-AnyLabelingcall venv\Scripts\activate.bat python anylabeling\app.py pause安装后验证 GPU启动 X-AnyLabeling 后在设置里选 GPU 推理或用命令行验证cd/d D:\ai\X-AnyLabeling venv\Scripts\activate.bat python-cimport torch; print(torch.cuda.is_available(), torch.cuda.get_device_name(0))