{"id":38730,"date":"2025-07-06T09:52:16","date_gmt":"2025-07-06T01:52:16","guid":{"rendered":"https:\/\/www.1ai.net\/?p=38730"},"modified":"2025-07-02T12:57:50","modified_gmt":"2025-07-02T04:57:50","slug":"comfyui-ai%e7%bb%98%e7%94%bb%e5%a4%aa%e6%85%a2%e6%80%8e%e4%b9%88%e5%8a%9e%ef%bc%9fgudacudnnpytorch%e5%ae%89%e8%a3%85%ef%bc%8c%e8%ae%a9%e4%bd%a0%e7%9a%84comfyui%e8%b5%b7%e9%a3%9e","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/38730.html","title":{"rendered":"What to do when comfyui AI painting is too slow?GUDA+cuDNN+pytorch installation to make your comfyUI take off!"},"content":{"rendered":"

\u6709\u4eba\u7684\u7535\u8111\u914d\u7f6e\u660e\u660e\u5f88\u9ad8\uff0c\u663e\u5361\u4e5f\u5f88\u597d\uff0c\u4f46\u5728\u505aAI\u7ed8\u753b<\/a>\u65f6\u901f\u5ea6\u5374\u5f88\u6162\uff0c\u90a3\u53ef\u80fd\u662f\u4f60\u6ca1\u6709\u5b89\u88c5CUDA\u548ccuDNN\u3002<\/p>\n

\u6211\u7684\u7535\u8111\u521a\u505a\u4e86\u7cfb\u7edf\uff0ccomfyUI\u4e5f\u91cd\u88c5\u4e86\uff0ccomfyUI \u6ca1\u88c5 CUDA,\u51fa\u56fe\u5c31\u53d8\u5f97\u5f88\u6162\u5f88\u6162\uff0c\u4eca\u5929\u5c31\u5e26\u5927\u5bb6\u628aCUDA\u3001cuDNN\u8fd8\u6709pytorch\u5b89\u88c5\u4e00\u4e0b\uff0c\u8ba9 comfyUI\u63d0\u4e2a\u901f\u3002<\/p>\n

\"43d674f1j00syr9ja0076d000u000irm\"<\/p>\n

\u5728\u6ca1\u5b89\u88c5\u524d\uff0c\u6211\u7528kontext\u8dd1\u4e86\u4e00\u5f20\u56fe\uff0c\u65f6\u95f4\u8fbe\u5230\u4e86285\u79d2\u3002<\/p>\n

\"be062c4bj00syr9j60003d0009u0010m\"<\/p>\n

\u4e00\u3001\u5b89\u88c5CUDA<\/p>\n

1\u3001\u5728\u5b89\u88c5\u524d\u8981\u5148\u786e\u5b9a\u4f60\u7684\u7535\u8111\u5b89\u88c5CUDA\u6700\u9ad8\u652f\u6301\u54ea\u4e2a\u7248\u672c<\/p>\n

\u9996\u5148\u5728\u7535\u8111\u4e0a\u6253\u5f00NVIDIA\u63a7\u5236\u9762\u677f<\/p>\n

\"e862d103j00syr9j8002gd000fm007xm\"<\/p>\n

\u70b9\u51fbNVIDIA\u63a7\u5236\u9762\u677f\u5de6\u4e0b\u89d2\u7684\u7cfb\u7edf\u4fe1\u606f–>\u7ec4\u4ef6<\/p>\n

\"ba1380a7j00syr9ja002md000qo00jrm\"<\/p>\n

\u627e\u5230\u4e86NVCGDA64.DLL\uff0c\u5c31\u8bf4\u660e\u6211\u4eec\u7684\u7535\u8111\u652f\u6301\u5b89\u88c5CUDA<\/p>\n

\u8bb0\u4f4f\u540e\u9762\u7684\u7248\u672c\u53f712.9.90\uff0c\u8fd9\u662f\u6211\u4eec\u7535\u8111\u652f\u6301\u7684\u6700\u9ad8\u7248\u672c\uff0c\u6211\u4eec\u53ea\u80fd\u5b89\u88c5\u5c0f\u4e8e\u7b49\u4e8e12.9.90\u7248\u672c\u7684CUDA\u3002<\/p>\n

2\u3001\u6839\u636epython\u786e\u5b9aCUDA\u7684\u7248\u672c\u53f7<\/p>\n

\u6253\u5f00PyTorch\u7684\u7f51\u9875(https:\/\/pytorch.org)\uff0c\u5411\u4e0b\u62c9\uff0c\u53ef\u4ee5\u770b\u5230python\u7684\u7248\u672c\u9009\u62e9<\/p>\n

\u56e0\u4e3a\u6211\u4eec\u7684CUDA\u6700\u9ad8\u652f\u630112.9.90\uff0c\u6240\u4ee5\u6211\u4eec\u53ef\u4ee5\u9009\u62e911.8\u300112.6\u548c12.8\u4e09\u4e2a\u7248\u672c\u7684CUDA\u3002<\/p>\n

\u8fd9\u91cc\u6211\u4eec\u9009\u62e912.8\u7248\u672c\u7684\u3002<\/p>\n

\"6544bd29j00syr9j90031d000qq00hjm\"<\/p>\n

3\u3001\u4e0b\u8f7dCUDA<\/p>\n

\u786e\u5b9a\u597dCUDA\u7684\u7248\u672c\u540e\uff0c\u8fdb\u5165CUDA\u7684\u4e0b\u8f7d\u9875\u9762\uff0c\u9009\u62e9\u6b63\u786e\u7684\u7248\u672c\u8fdb\u884c\u4e0b\u8f7d\uff1a<\/p>\n

\u7f51\u5740\uff1a<\/p>\n

https:\/\/developer.nvidia.com\/cuda-toolkit-archive<\/p>\n

\"6e9b11bej00syr9ja00gdd000u000r4m\"<\/p>\n

3\u3001\u5b89\u88c5CUDA<\/p>\n

\u4e0b\u8f7d\u597d\u540e\u8fdb\u884c\u5b89\u88c5<\/p>\n

\u56e0\u4e3a\u6211\u5df2\u7ecf\u5b89\u88c5\u8fc7Visual Studio\u4e86\uff0c\u6240\u4ee5\u6211\u8fd9\u91cc\u6ca1\u6709\u52fe\u9009\u5b83\u3002<\/p>\n

\u70b9\u51fb\u4e0b\u4e00\u6b65\uff0c\u4e00\u8def\u9ed8\u8ba4\uff0c\u76f4\u5230\u5b89\u88c5\u7ed3\u675f\u3002<\/p>\n

\"12d51b26j00syr9j80031d000gi00ccm\"<\/p>\n

\u5b89\u88c5\u597d\u540e\uff0c\u68c0\u67e5\u662f\u5426\u5b89\u88c5\u6210\u529f<\/p>\n

win+r \u6253\u5f00\u8fd0\u884c\u7a97\u53e3\uff0c\u8f93\u5165”cmd”\u6253\u5f00\u547d\u4ee4\u884c\u7a97\u53e3<\/p>\n

\u8f93\u5165”nvcc –version”\uff0c\u53ef\u4ee5\u67e5\u770bCUDA\u7248\u672c\u53f7\uff0c\u76ee\u524d\u7248\u672c\u53f7\u662fV12.8.61\uff0c\u8bf4\u660eCUDA\u5b89\u88c5\u6210\u529f<\/p>\n

\u518d\u8f93\u5165”set cuda”\u67e5\u770b\u73af\u5883\u53d8\u91cf\u662f\u5426\u5df2\u7ecf\u8bbe\u7f6e\u597d<\/p>\n

\"5dca8c98j00syr9j6001dd000kn00aom\"<\/p>\n

\u4f60\u8fd8\u53ef\u4ee5\u901a\u8fc7\u4e0b\u9762\u7684\u65b9\u6cd5\u67e5\u770b\u662f\u5426\u5df2\u7ecf\u8bbe\u7f6e\u597d\u73af\u5883\u53d8\u91cf\uff1a<\/p>\n

\u5728\u6211\u7684\u7535\u8111\u4e0a\u53f3\u51fb–>\u5c5e\u6027–>\u9ad8\u7ea7\u7cfb\u7edf\u8bbe\u7f6e<\/p>\n

\"25266232j00syr9j7001hd000u000iom\"<\/p>\n

\u70b9\u51fb\u73af\u5883\u53d8\u91cf\uff0c\u770b\u5230CUDA\u5df2\u7ecf\u6dfb\u52a0\u8fdb\u4e86\u7cfb\u7edf\u53d8\u91cf\u91cc\u9762<\/p>\n

\"56075bf9j00syr9ja00ald000u000gwm\"<\/p>\n

\u4e8c\u3001\u5b89\u88c5cuDNN<\/p>\n

CUDA\u8981\u548ccuDNN\u914d\u5408\u4e00\u5757\u4f7f\u7528<\/p>\n

cuDNN\u7684\u7248\u672c\u53f7\u5e94\u8be5\u7b49\u4e8e\u6216\u5c0f\u4e8eCUDA\u7684\u7248\u672c\u53f7<\/p>\n

cuDNN\u7f51\u5740\uff1a<\/p>\n

https:\/\/developer.nvidia.com\/rdp\/cudnn-archive#a-collapse897-120<\/p>\n

\u56e0\u4e3a\u6211\u4eec\u5b89\u88c5\u7684CUDA\u662f12.8\u7248\u672c\uff0c\u6240\u4ee5\u5bf9\u5e94\u7684cuDNN\u8981\u5b89\u88c58.9.7\u7248\u672c\u7684\u3002<\/p>\n

\"9f3205b4j00syr9ja00a1d000u000r4m\"<\/p>\n

\u70b9\u5f008.9.7\u7248\u672c\u7684\u94fe\u63a5\u540e\uff0c\u4e0b\u8f7d\u7b2c\u4e00\u4e2a\u672c\u5730\u5b89\u538b\u7f29\u5305\u3002<\/p>\n

\u5728\u4e0b\u8f7d\u524d\u9700\u8981\u767b\u9646\uff0c\u968f\u4fbf\u627e\u4e2a\u90ae\u7bb1\u6ce8\u518c\u4e00\u4e0b\u5c31\u53ef\u4ee5\u767b\u9646\u4e86\u3002<\/p>\n

\"3dd6c16aj00syr9ja0031d000ps00kom\"<\/p>\n

\u4e0b\u8f7d\u5b8c\u6210\u540e\u89e3\u538b\uff0c\u5f97\u5230\u4e00\u4e9b\u6587\u4ef6<\/p>\n

\"f2f07a27j00syr9j7000jd000gj006am\"<\/p>\n

\u590d\u5236\u8fd9\u4e9b\u6587\u4ef6\uff0c\u7136\u540e\u7c98\u8d34\u5230\u521a\u521a\u7684CUDA\u8def\u5f84\u4e0b\uff0c\u66ff\u6362\u539f\u6709\u6587\u4ef6<\/p>\n

CUDA\u7684\u9ed8\u8ba4\u5b89\u88c5\u8def\u5f84\u662f\uff1a<\/p>\n

C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8<\/p>\n

\"a446e398j00syr9j9001sd000lm00eum\"<\/p>\n

\u590d\u5236\u5b8c\u6210\u540e\uff0ccuDNN\u5b89\u88c5\u6210\u529f\u3002<\/p>\n

\u4e09\u3001\u5b89\u88c5PyTorch<\/p>\n

\u56de\u5230\u521a\u521aPyTorch\u7684\u9875\u9762\uff0c\u6ce8\u610f\u4e0b\u56fe\u4e2d\u5708\u51fa\u6765\u7684\u4e24\u4e2a\u90e8\u5206\uff0c\u4e0a\u9762\u5708\u8d77\u6765\u90e8\u5206\u7684\u610f\u601d\u662f Python \u5fc5\u987b\u89813.9\u6216\u4ee5\u540e\u7684\u7248\u672c<\/p>\n

\u5982\u679c\u4f60\u7684\u7535\u8111\u91cc\u6ca1\u6709 python\uff0c\u53ef\u4ee5\u5b89\u88c5\u4e00\u4e2a<\/p>\n

python:https:\/\/www.python.org\/<\/p>\n

\u5e95\u9762\u7684\u90e8\u5206\u662f pytorch \u7684\u5b89\u88c5\u547d\u4ee4\uff0c\u590d\u5236\u5b83<\/p>\n

\"a47a24a2j00syr9ja004ud000mr00nsm\"<\/p>\n

\u8c03\u51fa\u547d\u4ee4\u884c\u5de5\u5177\uff0c\u7c98\u8d34\u521a\u521a\u590d\u5236\u7684\u547d\u4ee4\uff0c\u6572\u56de\u8f66<\/p>\n

pip3 install torch torchvision torchaudio –index-url https:\/\/download.pytorch.org\/whl\/cu128<\/p>\n

\"a7eed619j00syr9j9000wd000u000gvm\"<\/p>\n

\u4e0b\u9762\u8010\u5fc3\u7b49\u5f85\u5b89\u88c5\u5b8c\u6210<\/p>\n

\"aa2e1dabj00syr9ja0064d000u000gwm\"<\/p>\n

\u73b0\u5728CUDA\u3001cuDNN\u8fd8\u6709pytorch\u5b89\u88c5\u5b8c\u6210\u4e86\uff0c\u8fd9\u65f6\u518d\u8fd0\u884c\u521a\u521a\u7684 kontext\u5de5\u4f5c\u6d41\uff0c\u65f6\u95f4\u660e\u663e\u51cf\u5c11\u4e86\u5f88\u591a\u3002<\/p>\n

\"834f6269j00syr9j60007d000fg001gm\"<\/p>\n

\u56db\u3001\u79cb\u53f6\u542f\u52a8\u5668\u7684\u5176\u5b83\u9ad8\u7ea7\u8bbe\u7f6e<\/p>\n

\u5176\u5b9e\u79cb\u53f6\u542f\u52a8\u5668\u91cc\u9762\u96c6\u6210\u4e86\u5f88\u591a\u5b9e\u7528\u7684\u529f\u80fd\uff0c\u4f1a\u4f7f\u7528\u5b83\uff0c\u80fd\u591f\u4e3a\u6211\u4eec\u5e26\u4e86\u5f88\u591a\u4fbf\u5229\u4e4b\u5904\u3002<\/p>\n

\u6309\u4e0b\u56fe\u8fdb\u5165\u4e13\u5bb6\u6a21\u5f0f\uff1a\u70b9\u51fb\u5de6\u4e0b\u89d2\u8bbe\u7f6e–>\u4e00\u822c\u8bbe\u7f6e–>\u4e13\u5bb6\u6a21\u5f0f<\/p>\n

\"fc773e06j00syr9ja004yd000u000irm\"<\/p>\n

1\u3001\u66f4\u6539\u9ed8\u8ba4\u6d4f\u89c8\u5668<\/p>\n

\u8fd9\u91cc\u5efa\u8bae\u4f7f\u7528\u8c37\u6b4c\u6d4f\u89c8\u5668\uff0c\u56e0\u4e3a\u8c37\u6b4c\u6d4f\u89c8\u5668\u6700\u7a33\u5b9a\uff0c\u5982\u679c\u4f60\u7684\u7535\u8111\u91cc\u6ca1\u6709\uff0c\u5efa\u8bae\u4e0b\u8f7d\u4e00\u4e2a\u3002<\/p>\n

\"36d57a66j00syr9ja004ud000u000irm\"<\/p>\n

2\u3001python\u8def\u5f84\u8986\u76d6<\/p>\n

\u6309\u4e0b\u56fe\u8bbe\u7f6e\u5b89\u88c5\u5305\u91cc\u7684python\u8def\u5f84<\/p>\n

\"d0ffc91aj00syr9ja005ud000u000irm\"<\/p>\n

2\u3001\u68c0\u67e5\u662f\u5426\u5b89\u88c5transformers\u548chuggingface<\/p>\n

\u56de\u5230\u79cb\u53f6comfyui\u7684python\u7684\u6587\u4ef6\u5939\uff0c\u5728\u5730\u5740\u680f\u5185\u6572\u5165 cmd \u7136\u540e\u56de\u8f66\uff0c\u8c03\u51fa\u547d\u4ee4\u884c<\/p>\n

\"a0cc6226j00syr9j9002cd000mk00i9m\"<\/p>\n

\u8f93\u5165\u547d\u4ee4\uff1apython.exe -m pip show transformers<\/p>\n

\u8fd9\u6761\u547d\u4ee4\u7528\u6765\u68c0\u67e5\u662f\u5426\u5b89\u88c5\u4e86 transformers<\/p>\n

\"4df2524ej00syr9j9002td000u000gwm\"<\/p>\n

\u6211\u8fd9\u4e2a\u662f\u5df2\u7ecf\u5b89\u88c5\u8fc7\u4e86\u3002 \u5982\u679c\u4f60\u7684\u6ca1\u6709\u5b89\u88c5\uff0c\u8f93\u5165\u4e0b\u9762\u547d\u4ee4\u8fdb\u884c\u5b89\u88c5\uff1a<\/p>\n

python.exe -m pip install transformers<\/p>\n

\u63a5\u7740\u518d\u68c0\u67e5huggingface\u6709\u6ca1\u6709\u5b89\u88c5\uff0c\u8f93\u5165\u4e0b\u9762\u547d\u4ee4\u8fdb\u884c\u68c0\u67e5<\/p>\n

python.exe -m pip show huggingface_hub<\/p>\n

\"1d5e5288j00syr9ja004md000u000gwm\"<\/p>\n

\u6211\u8fd9\u4e2a\u662f\u5df2\u7ecf\u5b89\u88c5\u8fc7\u4e86\u3002 \u5982\u679c\u4f60\u7684\u6ca1\u6709\u5b89\u88c5\uff0c\u8f93\u5165\u4e0b\u9762\u547d\u4ee4\u8fdb\u884c\u5b89\u88c5\uff1a<\/p>\n

python.exe -m pip install\u00a0huggingface_hub<\/p>\n

3\u3001\u7528\u540c\u6837\u7684\u65b9\u6cd5\u8bbe\u7f6egit\u8def\u5f84\u8986\u76d6<\/p>\n

\"a90e9eaaj00syr9ja005ld000u000irm\"<\/p>\n

4\u3001\u8bbe\u7f6e\u7f51\u7edc\u4ee3\u7406<\/p>\n

\u5728\u8bbe\u7f6e\u7f51\u7edc\u4ee3\u7406\u65f6\u8981\u786e\u4fdd\u7aef\u53e3\u53f7\u4e00\u81f4<\/p>\n

\u7136\u540e\u5f00\u542f\u4ee3\u7406\u540e\u9762\u7684\u56db\u4e2a\u5f00\u5173\uff0c\u540c\u65f6\u5173\u6389\u955c\u50cf\u5f00\u5173<\/p>\n

\"2d2fe245j00syr9ja004xd000u000irm\"<\/p>\n

\u5199\u5728\u6700\u540e<\/p>\n

\u5176\u5b9e\u4eca\u5929\u6211\u4eec\u8bb2\u7684GUDA\u3001PyTorch\u548cxformers\u5b89\u88c5\u65b9\u6cd5\u5728\u79cb\u53f6\u6574\u5408\u5305\u91cc\u53ef\u4ee5\u5f88\u8f7b\u677e\u5b8c\u6210\u3002<\/p>\n

\u5728\u79cb\u53f6\u6574\u5408\u5305\u91cc\u627e\u5230\u201c\u9ad8\u7ea7\u9009\u9879–>\u73af\u5883\u7ef4–>\u5b89\u88c5PyTorch\u201d\uff0c\u53ea\u8981\u70b9\u9009\u9700\u8981\u7684\u7248\u672c\uff0c\u5c31\u80fd\u65e0\u8111\u5b89\u88c5\u4e86\u3002<\/p>\n

\u5f53\u7136\uff0c\u4f60\u5728\u5b89\u88c5\u7684\u65f6\u5019\u6700\u597d\u6309\u6211\u4e0a\u9762\u8bb2\u7684\u65b9\u6cd5\u5148\u786e\u5b9a\u4f60\u7684\u7535\u8111\u652f\u6301\u54ea\u4e2a\u7248\u672c\u518d\u8fdb\u884c\u5b89\u88c5\u3002<\/p>\n

\"f6921890j00syr9ja005kd000u000ixm\"<\/p>\n

\u4ee5\u4e0a\u5c31\u662f\u4eca\u5929\u7684\u5168\u90e8\u5185\u5bb9\uff0c\u5e0c\u671b\u5bf9\u4f60\u6709\u6240\u5e2e\u52a9\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"

Someone with a very high computer setup and a good graphic card, but at a very slow pace when you're doing an AI painting, you probably didn't install CUDA and CUDNN. I've just finished my computer, and I've got a new system, and I've got no CUDA, and I've got a slow, slow picture, and I'm gonna need you to install CUDA, CUDNN, and pytorch today, so that we can move the cofyUI. Before it was installed, I ran a map with kontext, which was 285 seconds. I. Installation of CUDA 1 and determination of which version of the computer will be most supported by the CUDA before it is installed<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[144],"tags":[491,1989,6276],"collection":[],"class_list":["post-38730","post","type-post","status-publish","format-standard","hentry","category-baike","tag-ai","tag-comfyui"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/38730","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=38730"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/38730\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=38730"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=38730"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=38730"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=38730"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}