{"id":23,"date":"2026-05-07T21:58:25","date_gmt":"2026-05-07T12:58:25","guid":{"rendered":"https:\/\/material-ai-lab.com\/?p=23"},"modified":"2026-05-07T22:10:30","modified_gmt":"2026-05-07T13:10:30","slug":"%e3%80%90%e5%88%9d%e5%bf%83%e8%80%85%e5%90%91%e3%81%91%e3%80%91pytorch%e3%82%92windows%e3%81%a7gpu%e5%af%be%e5%bf%9c%e3%81%95%e3%81%9b%e3%82%8b%e7%92%b0%e5%a2%83%e6%a7%8b%e7%af%89%e6%96%b9%e6%b3%95","status":"publish","type":"post","link":"https:\/\/material-ai-lab.com\/?p=23","title":{"rendered":"\u3010\u521d\u5fc3\u8005\u5411\u3051\u3011PyTorch\u3092Windows\u3067GPU\u5bfe\u5fdc\u3055\u305b\u308b\u74b0\u5883\u69cb\u7bc9\u65b9\u6cd5\uff5cCUDA\u306e\u78ba\u8a8d\u304b\u3089\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u307e\u3067\u89e3\u8aac"},"content":{"rendered":"\n

PyTorch\u306f\u3001Windows\u3067\u306eGPU\u4f7f\u7528\u3092\u30cd\u30a4\u30c6\u30a3\u30d6\u306b\u30b5\u30dd\u30fc\u30c8\u3057\u3066\u3044\u308b<\/strong>\u3053\u3068\u306b\u52a0\u3048\u3001TensorFlow\u3084Chainer\u306a\u3069\u306eAI\u30e9\u30a4\u30d6\u30e9\u30ea\u3068\u6bd4\u3079\u3066\u3082\u73fe\u5728\u4e3b\u6d41\u306e\u9078\u629e\u80a2\u306e\u4e00\u3064<\/strong>\u3067\u3059\u3002\u305d\u306e\u305f\u3081\u3001\u3053\u308c\u304b\u3089\u6a5f\u68b0\u5b66\u7fd2\u3084\u6df1\u5c64\u5b66\u7fd2\u3092\u59cb\u3081\u308b\u65b9\u306b\u306f\u3001PyTorch\u3092\u304a\u3059\u3059\u3081\u3057\u307e\u3059\u3002<\/p>\n\n\n\n

\u4e00\u65b9\u3067\u3001\u521d\u3081\u3066\u74b0\u5883\u69cb\u7bc9\u3092\u884c\u3046\u5834\u5408\u3001GPU\u5bfe\u5fdc\u306e\u8a2d\u5b9a\u3067\u3064\u307e\u305a\u304d\u3084\u3059\u3044<\/strong>\u306e\u3082\u4e8b\u5b9f\u3067\u3059\u3002\u3053\u306e\u8a18\u4e8b\u3067\u306f\u3001Windows\u3067PyTorch\u3092GPU\u5bfe\u5fdc\u3067\u4f7f\u3046\u305f\u3081\u306e\u74b0\u5883\u69cb\u7bc9\u65b9\u6cd5<\/strong>\u3092\u3001\u521d\u5fc3\u8005\u5411\u3051\u306b\u5206\u304b\u308a\u3084\u3059\u304f\u8aac\u660e\u3057\u307e\u3059\u3002<\/p>\n\n\n\n

\u4eca\u56de\u5fc5\u8981\u306b\u306a\u308b\u306e\u306f\u3001\u4ee5\u4e0b\u306e2\u3064\u3067\u3059\u3002<\/p>\n\n\n\n

\u30fbCUDA
\u30fbPyTorch<\/p>\n\n\n\n

1. PyTorch\u306b\u5bfe\u5fdc\u3057\u3066\u3044\u308bCUDA\u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u3092\u78ba\u8a8d\u3059\u308b<\/h2>\n\n\n\n

CUDA\u306f\u3001NVIDIA\u304c\u958b\u767a\u3057\u305f\u3001GPU\u3092\u4f7f\u3063\u3066\u9ad8\u901f\u306a\u8a08\u7b97\u3092\u884c\u3046\u305f\u3081\u306e\u30d7\u30e9\u30c3\u30c8\u30d5\u30a9\u30fc\u30e0<\/strong>\u3067\u3059\u3002PyTorch\u3067GPU\u3092\u4f7f\u7528\u3059\u308b\u306b\u306f\u3001CUDA\u3078\u306e\u5bfe\u5fdc\u3092\u78ba\u8a8d\u3059\u308b\u5fc5\u8981\u304c\u3042\u308a\u307e\u3059\u3002<\/p>\n\n\n\n

\u307e\u305a\u306f\u3001\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u305f\u3044PyTorch\u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u304c\u3001\u3069\u306eCUDA\u30d0\u30fc\u30b8\u30e7\u30f3\u306b\u5bfe\u5fdc\u3057\u3066\u3044\u308b\u304b<\/strong>\u3092\u516c\u5f0f\u30b5\u30a4\u30c8\u304b\u3089\u78ba\u8a8d\u3057\u307e\u3057\u3087\u3046\u30022026\u5e744\u6708\u6642\u70b9\u3067\u306f\u3001CUDA 12.6\u300112.8\u300113.0\u306b\u5bfe\u5fdc\u3057\u3066\u3044\u307e\u3059\u3002<\/p>\n\n\n\n

\u307e\u305f\u3001Windows\u7248PyTorch\u3067\u306fAMD\u88fdGPU\u306f\u5229\u7528\u3067\u304d\u306a\u3044<\/strong>\u305f\u3081\u3001NVIDIA\u88fdGPU\u304c\u5fc5\u8981\u306b\u306a\u308b\u70b9\u306b\u3082\u6ce8\u610f\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n\n\n\n

\"\"\/<\/a><\/figure>\n\n\n\n
\nhttps:\/\/pytorch.org\/get-started\/locally\n<\/div><\/figure>\n\n\n\n

2. CUDA Toolkit\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b<\/h2>\n\n\n\n

\u4eca\u56de\u306f\u3001CUDA 13.0<\/strong>\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b\u4f8b\u3067\u8aac\u660e\u3057\u307e\u3059\u3002CUDA\u306b\u306f\u8907\u6570\u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u304c\u3042\u308b\u305f\u3081\u3001PyTorch\u306b\u5bfe\u5fdc\u3057\u305f\u30d0\u30fc\u30b8\u30e7\u30f3\u3092\u9078\u3076\u3053\u3068\u304c\u91cd\u8981<\/strong>\u3067\u3059\u3002<\/p>\n\n\n\n

Google\u3067\u300cCUDA 13.0\u300d\u3068\u691c\u7d22\u3057\u3001CUDA\u306e\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u30da\u30fc\u30b8\u306b\u30a2\u30af\u30bb\u30b9\u3057\u307e\u3059\u3002\u305d\u3053\u3067\u3001\u4f7f\u7528\u3057\u3066\u3044\u308b\u74b0\u5883\u306b\u5408\u308f\u305b\u3066\u4ee5\u4e0b\u3092\u9078\u629e\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n\n\n\n

\u30fbOS\uff1aWindows
\u30fbArchitecture\uff1a\u4f7f\u7528\u4e2d\u306e\u74b0\u5883\u306b\u5408\u3063\u305f\u3082\u306e
\u30fbWindows Version\uff1a\u4f7f\u7528\u4e2d\u306eWindows\u306e\u30d0\u30fc\u30b8\u30e7\u30f3
\u30fbInstaller Type\uff1aexe (local)<\/strong><\/p>\n\n\n\n

\u305d\u306e\u5f8c\u3001CUDA Toolkit Installer<\/strong>\u3092\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3057\u307e\u3059\u3002\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u306f\u3001\u753b\u9762\u306e\u6848\u5185\u306b\u5f93\u3063\u3066\u9032\u3081\u308c\u3070\u554f\u984c\u3042\u308a\u307e\u305b\u3093\u3002<\/p>\n\n\n\n

\"\"\/<\/figure>\n\n\n\n
\nhttps:\/\/developer.nvidia.com\/cuda-13-0-0-download-archive\n<\/div><\/figure>\n\n\n\n

3. PyTorch\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b<\/h2>\n\n\n\n

\u6b21\u306b\u3001PyTorch\u306e\u516c\u5f0f\u30b5\u30a4\u30c8\u3078\u79fb\u52d5\u3057\u307e\u3059\u3002\u30b5\u30a4\u30c8\u4e0a\u3067\u3001\u81ea\u5206\u306e\u74b0\u5883\u306b\u5408\u3063\u305f\u8a2d\u5b9a<\/strong>\u3092\u9078\u629e\u3057\u3066\u304f\u3060\u3055\u3044\u3002\u8a2d\u5b9a\u304c\u7d42\u308f\u308b\u3068\u3001**\u300cRun this Command\u300d**\u3068\u3044\u3046\u5f62\u3067\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u7528\u30b3\u30de\u30f3\u30c9\u304c\u8868\u793a\u3055\u308c\u307e\u3059\u3002<\/p>\n\n\n\n

\u305d\u306e\u30b3\u30de\u30f3\u30c9\u3092\u30b3\u30d4\u30fc\u3057\u3001\u30b3\u30de\u30f3\u30c9\u30e9\u30a4\u30f3\u306b\u8cbc\u308a\u4ed8\u3051\u3066\u5b9f\u884c\u3059\u308c\u3070\u3001PyTorch\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3067\u304d\u307e\u3059\u3002<\/p>\n\n\n\n

\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u5b8c\u4e86\u5f8c\u306f\u3001\u4ee5\u4e0b\u306e\u30b3\u30de\u30f3\u30c9\u3092\u5b9f\u884c\u3057\u3066\u3001GPU\u304c\u6b63\u3057\u304f\u4f7f\u3048\u308b\u304b\u78ba\u8a8d\u3057\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n

\"\"\/<\/a><\/figure>\n\n\n\n
\nhttps:\/\/pytorch.org\/get-started\/locally\n<\/div><\/figure>\n\n\n\n

\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u5b8c\u4e86\u5f8c\u306f\u3001\u4ee5\u4e0b\u306e\u30b3\u30de\u30f3\u30c9\u3092\u5b9f\u884c\u3057\u3066\u3001GPU\u304c\u6b63\u3057\u304f\u4f7f\u3048\u308b\u304b\u78ba\u8a8d\u3057\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n

python -c “import torch; print(torch.device(‘cuda’ if torch.cuda.is_available() else ‘cpu’))”<\/p>\n\n\n\n

\u3053\u306e\u30b3\u30de\u30f3\u30c9\u3092\u5b9f\u884c\u3057\u305f\u3068\u304d\u3001cuda<\/strong>\u3068\u8868\u793a\u3055\u308c\u308c\u3070\u3001GPU\u304c\u4f7f\u7528\u53ef\u80fd\u306a\u72b6\u614b\u3067\u3059\u3002
\u4e00\u65b9\u3001cpu<\/strong>\u3068\u8868\u793a\u3055\u308c\u305f\u5834\u5408\u306f\u3001GPU\u304c\u6709\u52b9\u306b\u306a\u3063\u3066\u3044\u307e\u305b\u3093\u3002<\/p>\n\n\n\n

\u305d\u306e\u5834\u5408\u306f\u3001\u4ee5\u4e0b\u306e\u3088\u3046\u306a\u70b9\u3092\u78ba\u8a8d\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n\n\n\n

\u30fbCUDA\u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u304cPyTorch\u306b\u5bfe\u5fdc\u3057\u3066\u3044\u308b\u304b
\u30fbPyTorch\u3092GPU\u5bfe\u5fdc\u7248\u3067\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u3066\u3044\u308b\u304b
\u30fbNVIDIA\u88fdGPU\u3092\u4f7f\u7528\u3057\u3066\u3044\u308b\u304b<\/p>\n\n\n\n

4. \u307e\u3068\u3081<\/h2>\n\n\n\n

\u4eca\u56de\u306f\u3001Windows\u3067GPU\u5bfe\u5fdc\u306ePyTorch\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b\u65b9\u6cd5<\/strong>\u3092\u7d39\u4ecb\u3057\u307e\u3057\u305f\u3002<\/p>\n\n\n\n

\u624b\u9806\u306e\u6d41\u308c\u306f\u3001\u4ee5\u4e0b\u306e\u3068\u304a\u308a\u3067\u3059\u3002<\/p>\n\n\n\n

\u30fbPyTorch\u306b\u5bfe\u5fdc\u3057\u3066\u3044\u308bCUDA\u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u3092\u78ba\u8a8d\u3059\u308b
\u30fbCUDA Toolkit\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b
\u30fbPyTorch\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b
\u30fb\u30b3\u30de\u30f3\u30c9\u3067GPU\u304c\u4f7f\u3048\u308b\u304b\u78ba\u8a8d\u3059\u308b<\/p>\n\n\n\n

PyTorch\u3092GPU\u5bfe\u5fdc\u3067\u4f7f\u3048\u308b\u3088\u3046\u306b\u306a\u308b\u3068\u3001\u5b66\u7fd2\u901f\u5ea6\u304c\u5927\u304d\u304f\u5411\u4e0a\u3057\u3001\u6a5f\u68b0\u5b66\u7fd2\u3084\u6df1\u5c64\u5b66\u7fd2\u306e\u958b\u767a\u304c\u9032\u3081\u3084\u3059\u304f\u306a\u308a\u307e\u3059\u3002\u3053\u308c\u304b\u3089Windows\u74b0\u5883\u3067AI\u958b\u767a\u3092\u59cb\u3081\u308b\u65b9\u306f\u3001\u307e\u305a\u3053\u306e\u74b0\u5883\u69cb\u7bc9\u304b\u3089\u9032\u3081\u3066\u307f\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"

PyTorch\u306f\u3001Windows\u3067\u306eGPU\u4f7f\u7528\u3092\u30cd\u30a4\u30c6\u30a3\u30d6\u306b\u30b5\u30dd\u30fc\u30c8\u3057\u3066\u3044\u308b\u3053\u3068\u306b\u52a0\u3048\u3001TensorFlow\u3084Chainer\u306a\u3069\u306eAI\u30e9\u30a4\u30d6\u30e9\u30ea…<\/p>\n","protected":false},"author":1,"featured_media":19,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[4,7],"tags":[],"class_list":["post-23","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai","category-python"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=\/wp\/v2\/posts\/23","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=23"}],"version-history":[{"count":3,"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=\/wp\/v2\/posts\/23\/revisions"}],"predecessor-version":[{"id":29,"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=\/wp\/v2\/posts\/23\/revisions\/29"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=\/wp\/v2\/media\/19"}],"wp:attachment":[{"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=23"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=23"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/material-ai-lab.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=23"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}