{"id":55758,"date":"2026-08-16T09:45:01","date_gmt":"2026-08-16T01:45:01","guid":{"rendered":"https:\/\/www.1ai.net\/?p=55758"},"modified":"2026-08-11T09:40:58","modified_gmt":"2026-08-11T01:40:58","slug":"comfyui-%e4%bb%8e%e5%ae%89%e8%a3%85%e5%85%a5%e9%97%a8%e5%88%b0%e7%b2%be%e9%80%9a%ef%bc%8c%e6%8a%a5%e9%94%99%e8%87%aa%e6%95%91%ef%bc%88%e8%af%bb%e6%87%82%e7%ba%a2%e5%ad%97-%e5%b8%b8%e8%a7%81%e9%94%99","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/55758.html","title":{"rendered":"ComfyUI from installation entry to mastery, reporting faulty self-help (reads red + common error step resolution)"},"content":{"rendered":"<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-55802\" title=\"63ee2d34j00tjl0ne00ncd000ow00exp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/08\/63ee2d34j00tjl0ne00ncd000ow00exp.jpg\" alt=\"63ee2d34j00tjl0ne00ncd000ow00exp\" width=\"896\" height=\"537\" \/><\/p>\n<p>\u4e0a\u7bc7\uff08\u7b2c06\u7bc7\uff09<a href=\"https:\/\/www.1ai.net\/55757.html\">ComfyUI \u4ece\u5b89\u88c5\u5165\u95e8\u5230\u7cbe\u901a\uff0c\u51fa\u56fe\u8d28\u91cf\u63d0\u5347\uff08\u5206\u8fa8\u7387\/\u6b65\u6570\/\u91c7\u6837\u5668\/CFG\/\u9ad8\u6e05\u653e\u5927 \u00a0<\/a>\u53c2\u6570\u8c03\u597d\u4e86\uff0c\u753b\u8d28\u4e5f\u4e0a\u6765\u4e86\u3002\u4f46\u73a9 AI \u7ed8\u56fe\uff0c\u8c01\u8fd8\u6ca1\u89c1\u8fc7\u6ee1\u5c4f\u7ea2\u5b57\uff1f<\/p>\n<p>\u6211\u7b2c\u4e00\u6b21\u9047\u62a5\u9519\u90a3\u4f1a\u513f\uff0c\u76f4\u63a5\u61f5\u4e86\u2014\u2014\u4e00\u5927\u6bb5\u82f1\u6587\u5237\u4e0b\u6765\uff0c\u770b\u7740\u50cf\u7a0b\u5e8f\u5d29\u4e86\uff0c\u5dee\u70b9\u628a\u8f6f\u4ef6\u5378\u4e86\u91cd\u88c5\u3002\u540e\u6765\u624d\u53d1\u73b0\uff0c<a href=\"https:\/\/www.1ai.net\/en\/tag\/comfyui\" title=\"\u3010\u67e5\u770b\u542b\u6709[ComfyUI]\u6807\u7b7e\u7684\u6587\u7ae0\u3011\" target=\"_blank\" rel=\"noopener\">ComfyUI<\/a> \u7684\u62a5\u9519\u5176\u5b9e\u662f&#8221;\u75c7\u72b6\u4e66&#8221;\uff0c<strong>\u770b\u61c2\u7ea2\u5b57\u6700\u540e\u4e00\u884c\uff0c\u5927\u90e8\u5206\u95ee\u9898\u81ea\u5df1\u5c31\u80fd\u89e3\u51b3<\/strong>\u3002\u8fd9\u7bc7\u628a\u65b0\u624b\u6700\u5e38\u649e\u89c1\u7684\u51e0\u4e2a\u7ea2\u5b57\u9010\u4e2a\u62c6\uff1a\u73b0\u8c61\u2192\u539f\u56e0\u2192\u600e\u4e48\u4fee\uff0c\u6bcf\u4e2a\u90fd\u5e26\u6211\u8e29\u8fc7\u7684\u5751\u3002\u5efa\u8bae\u6536\u85cf\uff0c\u54ea\u5929\u649e\u4e0a\u4e86\u76f4\u63a5\u7ffb\u3002<\/p>\n<p><strong>\u62a5\u95191\uff1a\u8282\u70b9\u53d8\u7ea2\u8272\uff08Node type not found\uff09<\/strong><\/p>\n<p><strong>\u75c7\u72b6\uff1a<\/strong>\u6253\u5f00 ComfyUI \u6216\u5bfc\u5165\u5de5\u4f5c\u6d41\u540e\uff0c\u67d0\u4e2a\u8282\u70b9\u53d8\u7ea2\u6846\uff0c\u5199\u7740 &#8220;Node xxx failed to load&#8221;\u3002<br \/>\n<strong>\u539f\u56e0\uff1a<\/strong>\u5de5\u4f5c\u6d41\u7528\u4e86\u4f60\u6ca1\u88c5\u7684\u63d2\u4ef6\uff08\u81ea\u5b9a\u4e49\u8282\u70b9\uff09\u3002<br \/>\n<strong>\u89e3\u51b3\uff08\u6309\u987a\u5e8f\u6765\uff09\uff1a<\/strong><br \/>\n1. \u70b9\u53f3\u4fa7 &#8220;Manager&#8221;\uff08\u524d\u63d0\u88c5\u4e86 ComfyUI-Manager\uff0c\u6ca1\u88c5\u770b\u7b2c03\u7bc7\uff09<br \/>\n2. \u70b9 &#8220;Install Missing Custom Nodes&#8221;<br \/>\n3. \u7cfb\u7edf\u81ea\u52a8\u641c\u7f3a\u5931\u8282\u70b9\uff0c\u70b9 &#8220;Install&#8221;<br \/>\n4. \u91cd\u542f ComfyUI<br \/>\n\u8fd8\u7ea2\uff1aManager \u91cc\u641c\u8282\u70b9\u540d\u624b\u52a8\u88c5\uff1b\u6ca1 Manager \u5c31\u8fdb\u00a0 custom_nodes \u76ee\u5f55 git clone\u00a0\u62c9\u4ed3\u5e93\u3001pip install -r requirements.txt\u3001\u91cd\u542f\u3002<\/p>\n<p><strong>\u8e29\u5751\u7ecf\u5386\uff1a<\/strong>\u6211\u7b2c\u4e00\u6b21\u5bfc\u5165\u4e00\u4e2a ControlNet \u5de5\u4f5c\u6d41\uff0c\u7ea2\u8272\u8282\u70b9\u4e03\u516b\u4e2a\uff0c\u8fd8\u4ee5\u4e3a ComfyUI \u574f\u4e86\uff0c\u91cd\u88c5\u4e24\u904d\u624d\u53d1\u73b0\u662f\u6ca1\u88c5\u63d2\u4ef6\u3002<\/p>\n<p><strong>\u62a5\u95192\uff1a\u6a21\u578b\u627e\u4e0d\u5230\uff08Model not found \/ \u4e0b\u62c9\u7a7a\uff09<\/strong><\/p>\n<p><strong>\u75c7\u72b6\uff1a<\/strong>Load Checkpoint \u4e0b\u62c9\u662f\u7a7a\u7684\uff0c\u6216\u7ea2\u5b57\u00a0\u3002<br \/>\n<strong>\u539f\u56e0\uff1a<\/strong>\u6a21\u578b\u6ca1\u4e0b\u8f7d\uff0c\u6216\u653e\u9519\u4f4d\u7f6e\u3001\u6587\u4ef6\u540d\u5e26\u4e2d\u6587\/\u7a7a\u683c\u3002<br \/>\n<strong>\u89e3\u51b3\uff1a<\/strong>\u770b\u62a5\u9519\u91cc\u7684\u6a21\u578b\u540d \u2192 \u53bb\u6a21\u578b\u7ad9\uff08Civitai\u3001Hugging Face\uff09\u4e0b\u8f7d \u2192 \u653e\u8fdb\u5199\u6b7b\u7684\u76ee\u5f55 \u2192 \u70b9 Refresh \u6216\u91cd\u542f\u3002<\/p>\n<table>\n<tbody>\n<tr>\n<td>\n<section>\u6a21\u578b\u7c7b\u578b<\/section>\n<\/td>\n<td>\n<section>\u5b58\u653e\u4f4d\u7f6e<\/section>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<section>\u5e95\u6a21 Checkpoint<\/section>\n<\/td>\n<td><code>models\/checkpoints\/<\/code><\/td>\n<\/tr>\n<tr>\n<td>\n<section>LoRA<\/section>\n<\/td>\n<td><code>models\/loras\/<\/code><\/td>\n<\/tr>\n<tr>\n<td>\n<section>VAE<\/section>\n<\/td>\n<td><code>models\/vae\/<\/code><\/td>\n<\/tr>\n<tr>\n<td>\n<section>ControlNet<\/section>\n<\/td>\n<td><code>models\/controlnet\/<\/code><\/td>\n<\/tr>\n<tr>\n<td>\n<section>GGUF \u4e3b\u6a21\u578b<\/section>\n<\/td>\n<td><code>models\/unet\/<\/code><\/p>\n<section>\uff08\u6216 diffusion_models\/\uff09<\/section>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<section>\u91cf\u5316\u6587\u672c\u7f16\u7801\u5668<\/section>\n<\/td>\n<td><code>models\/text_encoders\/<\/code><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p>\u5c0f\u8d34\u58eb\uff1a\u6587\u4ef6\u540d\u4e0d\u5fc5\u548c\u62a5\u9519\u5b8c\u5168\u4e00\u81f4\uff0c\u5e95\u6a21\u7c7b\u578b\u5bf9\u5c31\u884c\u3002\u540e\u7f00\u8ba4 .safetensors\u00a0\/.ckpt\u00a0\uff0c\u82f1\u6587\u547d\u540d\u3001\u522b\u5e26\u7a7a\u683c\u3002<\/p>\n<p><strong>\u8e29\u5751\u7ecf\u5386\uff1a<\/strong>\u6211\u6709\u6b21\u6587\u4ef6\u540d\u5e26\u4e86\u4e2a\u7a7a\u683c\uff0c\u6b7b\u6d3b\u8bfb\u4e0d\u51fa\u6765\uff0c\u6539\u6389\u7a7a\u683c\u79d2\u597d\u2014\u2014\u8fd9\u79cd\u4f4e\u7ea7\u5751\u6700\u78e8\u4eba\u3002<\/p>\n<p><strong>\u62a5\u95193\uff1a\u7206\u663e\u5b58\uff08CUDA out of memory\uff09<\/strong><\/p>\n<p><strong>\u75c7\u72b6\uff1a<\/strong>\u70b9\u751f\u6210\u540e\u5f39 CUDA out of memory \uff0c\u5361\u6b7b\u6216\u95ea\u9000\u3002<br \/>\n<strong>\u539f\u56e0\uff1a<\/strong>\u663e\u5b58\u4e0d\u591f\u2014\u2014\u5206\u8fa8\u7387\u592a\u9ad8 \/ \u6a21\u578b\u592a\u5927 \/ batch \u592a\u5927\uff0c\u4e09\u9009\u4e00\u6216\u53e0\u52a0\u3002<br \/>\n<strong>\u89e3\u51b3\uff08\u4ece\u7b80\u5355\u5230\u590d\u6742\uff0c\u4e00\u6b21\u8bd5\u4e00\u9879\uff09\uff1a<\/strong><br \/>\n1. \u964d\u5206\u8fa8\u7387\uff0c512\u00d7512 \u8d77\u6b65\uff0c\u522b\u4e00\u4e0a\u6765\u5c31 1024<br \/>\n2. \u7528 SD1.5 \u66ff\u4ee3 SDXL\uff08SDXL \u663e\u5b58\u8981\u6c42\u9ad8\u5f88\u591a\uff09<br \/>\n3. \u542f\u52a8\u52a0 &#8211;lowvram\uff08\u6216&#8211;medvram\/ &#8211;novram\uff09<br \/>\n4. \u5173\u6389\u5360\u663e\u5b58\u7684\u7a0b\u5e8f\uff08\u6d4f\u89c8\u5668\u6807\u7b7e\u9875\u3001\u5fae\u4fe1\uff09<br \/>\n5. \u91cf\u5316\u6a21\u578b\uff0c\u7528 GGUF \/ FP8 \u91cf\u5316\u7248\uff08\u7b2c17\u7bc7\u4e13\u8bb2\uff09<\/p>\n<table>\n<tbody>\n<tr>\n<td>\n<section>\u663e\u5b58<\/section>\n<\/td>\n<td>\n<section>\u5b89\u5168\u5206\u8fa8\u7387<\/section>\n<\/td>\n<td>\n<section>\u63a8\u8350\u6a21\u578b<\/section>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<section>6GB<\/section>\n<\/td>\n<td>\n<section>512\u00d7512<\/section>\n<\/td>\n<td>\n<section>SD1.5<\/section>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<section>8GB<\/section>\n<\/td>\n<td>\n<section>512~768<\/section>\n<\/td>\n<td>\n<section>SD1.5 \/ SDXL\uff08\u91cf\u5316\u7248\uff09<\/section>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<section>12GB<\/section>\n<\/td>\n<td>\n<section>1024\u00d71024<\/section>\n<\/td>\n<td>\n<section>SDXL \/ Flux\uff08GGUF\uff09<\/section>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<section>16GB+<\/section>\n<\/td>\n<td>\n<section>\u968f\u4fbf<\/section>\n<\/td>\n<td>\n<section>\u5565\u90fd\u80fd\u8dd1<\/section>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><strong>\u8e29\u5751\u7ecf\u5386\uff1a<\/strong>\u6211\u5f53\u521d 8G \u663e\u5b58\uff0c\u7b2c\u4e00\u6b21\u8dd1 SDXL \u76f4\u63a5\u8bbe 1536\u00d71536\uff0c\u8fde\u62a5\u9519\u90fd\u6ca1\u5f39\u5c31\u95ea\u9000\uff0c\u540e\u6765\u624d\u77e5\u9053 1024\u00d71024 \u624d\u662f\u5b89\u5168\u8303\u56f4\u3002VAE \u89e3\u7801\u7206\uff0c\u6362 VAE Decode (Tiled)\u00a0\u6216\u542f\u52a8\u52a0&#8211;cpu-vae\u00a0\u3002<\/p>\n<p><strong>\u62a5\u95194\uff1a\u9ed1\u56fe \/ \u7cca\u56fe\uff08\u5e38\u5e38\u6ca1\u7ea2\u5b57\uff09<\/strong><\/p>\n<p><strong>\u75c7\u72b6\uff1a<\/strong>\u8dd1\u5b8c\u4e86\uff0c\u56fe\u5168\u9ed1\u3001\u53d1\u7070\u6216\u8272\u504f\uff0c\u8fd8\u6ca1\u62a5\u9519\u3002<br \/>\n<strong>\u539f\u56e0\uff1a<\/strong>VAE \u6ca1\u52a0\u8f7d\u5bf9\uff0c\u6216 CFG \u8fc7\u9ad8\u6570\u503c\u6ea2\u51fa\uff08NaN\uff09\u3002<br \/>\n<strong>\u89e3\u51b3\uff1a<\/strong><br \/>\n1. \u68c0\u67e5\u00a0 VAE Decode \u662f\u5426\u63a5\u4e86CheckpointLoaderSimple\u00a0\u00a0\u7684 VAE \u8f93\u51fa<br \/>\n2. \u5355\u72ec VAE \u6587\u4ef6\u8981\u5339\u914d\u5e95\u6a21\uff08SDXL \u914d SDXL \u7684 VAE\uff0cFlux \u7528\u81ea\u5bb6 VAE\uff09<br \/>\n3. CFG \u62c9\u56de\u5b89\u5168\u533a\uff08SDXL\/SD1.5 \u7ea6 6\uff5e8\uff1bFlux \u7528 FluxGuidance \u7ea6 1\uff5e3.5\uff0cKSampler \u91cc CFG \u8bbe 1\uff09<br \/>\n4. \u8001\u663e\u5361\uff08GTX16 \u7cfb\uff09\u542f\u52a8\u52a0 &#8211;force-fp32<\/p>\n<p><strong>\u8e29\u5751\u7ecf\u5386\uff1a<\/strong>\u6211\u9ed1\u56fe\u90a3\u6b21\uff0c\u7b2c\u4e00\u53cd\u5e94\u662f\u663e\u5361\u574f\u4e86\uff0c\u6298\u817e\u534a\u5929\uff0c\u6700\u540e\u53d1\u73b0\u662f VAE \u8282\u70b9\u6ca1\u8fde\u4e0a\u2014\u2014\u865a\u60ca\u4e00\u573a\u3002<\/p>\n<p><strong>\u62a5\u95195\uff1a\u751f\u6210\u7279\u522b\u6162<\/strong><\/p>\n<p><strong>\u75c7\u72b6\uff1a<\/strong>\u4e00\u5f20\u56fe\u8dd1\u597d\u51e0\u5206\u949f\u751a\u81f3\u51e0\u5341\u5206\u949f\u3002<br \/>\n<strong>\u539f\u56e0\uff1a<\/strong>\u91c7\u6837\u5668\u4e0d\u5f53\u3001\u6b65\u6570\u592a\u591a\u3001\u6a21\u578b\u592a\u5927\uff0c\u6216\u66f4\u5751\u7684\u2014\u2014\u6839\u672c\u5728\u7528 CPU \u8dd1\u3002<br \/>\n<strong>\u89e3\u51b3\uff1a<\/strong><br \/>\n1. \u6362\u91c7\u6837\u5668\uff1aeuler\u00a0\u6700\u5feb\uff0c dpmpp_2m \u53c8\u5feb\u53c8\u597d<br \/>\n2. \u964d\u6b65\u6570\uff1a\u65e5\u5e38 20 \u6b65\u591f\uff0c\u522b\u8bbe 50<br \/>\n3. \u4f4e\u5206\u8fa8\u7387\u5148\u6d4b\u6784\u56fe\uff0c\u518d\u8dd1\u9ad8\u53c2<br \/>\n4. \u52a0\u542f\u52a8\u53c2\u6570 &#8211;use-split-cross-attention\u00a0 \u63d0\u901f<br \/>\n5. \u770b\u7ec8\u7aef\u662f\u4e0d\u662f\u7528 CPU \u8dd1\uff08\u4f1a\u663e\u793a\uff09\uff0c\u662f\u7684\u8bdd\u73af\u5883\u914d\u9519\u4e86<br \/>\n\u65e5\u5e38\u63a8\u8350\uff1adpmpp_2m\u00a0+ 20 \u6b65 + karras\u3002<\/p>\n<p><strong>\u62a5\u95196\uff1a\u961f\u5217\u5361\u6b7b \/ \u4e0d\u51fa\u56fe<\/strong><\/p>\n<p><strong>\u75c7\u72b6\uff1a<\/strong>\u70b9 Queue Prompt \u6ca1\u53cd\u5e94\uff0c\u8fdb\u5ea6\u4e0d\u52a8\u3002<br \/>\n<strong>\u539f\u56e0\uff1a<\/strong>\u8282\u70b9\u8f93\u5165\u6ca1\u63a5\u597d\u5361\u4f4f\u3001\u7b2c\u4e09\u65b9\u8282\u70b9\u7be1\u6539\u961f\u5217\u3001\u6216\u663e\u5b58\u5360\u6ee1\u3002<br \/>\n<strong>\u89e3\u51b3\uff1a<\/strong><br \/>\n1. \u68c0\u67e5 KSampler \u5404\u8f93\u5165\u53e3\u63a5\u597d<br \/>\n2. \u70b9\u961f\u5217\u56fe\u6807\u6279\u91cf\u6e05\u7a7a\uff08\u53d6\u6d88\u5168\u90e8\uff09<br \/>\n3. \u786c\u5237\u65b0 Ctrl+Shift+R \u5e76\u91cd\u542f ComfyUI<br \/>\n4. \u6000\u7591\u67d0\u8282\u70b9\u4f5c\u5996\uff0c\u628a\u5b83\u7684\u6587\u4ef6\u5939\u5728 custom_nodes\u00a0\u6539\u540d\u5907\u4efd\uff0c\u91cd\u542f\u6392\u67e5<\/p>\n<p><strong>\u8e29\u5751\u7ecf\u5386\uff1a<\/strong>\u6709\u6b21\u6211\u4ee5\u4e3a\u8f6f\u4ef6\u5d29\u4e86\u51c6\u5907\u91cd\u88c5\uff0c\u7ed3\u679c\u662f\u4e0a\u4e00\u6761\u961f\u5217\u5361\u4f4f\u6ca1\u6e05\uff0c\u6e05\u7a7a\u79d2\u597d\u3002<\/p>\n<p><strong>\u62a5\u95197\uff1a\u6253\u4e0d\u5f00 \/ \u5361\u5728 loading<\/strong><\/p>\n<p><strong>\u75c7\u72b6\uff1a<\/strong>\u53cc\u51fb\u542f\u52a8\u540e\uff0c\u6d4f\u89c8\u5668\u00a0 127.0.0.1:8188 \u4e00\u76f4\u8f6c\u5708\u3002<br \/>\n<strong>\u539f\u56e0\uff1a<\/strong>\u6d4f\u89c8\u5668\u7f13\u5b58\u95ee\u9898\uff0c\u6216\u7aef\u53e3\u88ab\u5360\u7528\u3002<br \/>\n<strong>\u89e3\u51b3\uff1a<\/strong><br \/>\n1. \u6e05\u6d4f\u89c8\u5668\u7f13\u5b58\u6216\u7528\u65e0\u75d5\u6a21\u5f0f\u6253\u5f00<br \/>\n2. \u6362\u7aef\u53e3\uff1a\u542f\u52a8\u52a0 &#8211;port 8189<br \/>\n3. \u6740\u8fdb\u7a0b\uff1a\u4efb\u52a1\u7ba1\u7406\u5668\u5173\u6389\u6240\u6709 Python \u8fdb\u7a0b\uff0c\u91cd\u65b0\u542f\u52a8<\/p>\n<p><strong>\u62a5\u95198\uff08\u8fdb\u9636\uff09\uff1atorch \/ CUDA \u4e0d\u5339\u914d<\/strong><\/p>\n<p>\u8fd9\u4e2a\u65b0\u624b\u5c11\u89c1\uff0c\u4f46\u4e00\u65e6\u9047\u5230\u5f88\u61f5\u3002<br \/>\n<strong>\u75c7\u72b6\uff1a<\/strong>\u7ea2\u5b57 Torch not compiled with CUDA enabled\u00a0\u3002<br \/>\n<strong>\u539f\u56e0\uff1a<\/strong>PyTorch \u88c5\u6210 CPU \u7248\uff0c\u6216\u7248\u672c\u548c\u9a71\u52a8\u4e0d\u5bf9\u5e94\u3002<br \/>\n<strong>\u89e3\u51b3\uff1a<\/strong><br \/>\n1.\u00a0 nvidia-smi \u770b\u9876\u90e8 CUDA Version<br \/>\n2. \u5728 ComfyUI \u81ea\u5e26 Python \u91cc pip uninstall -y torch torchvision torchaudio<br \/>\n3. \u91cd\u88c5\u5bf9\u5e94 cu \u7248\uff08cu126\/cu128\/cu130 \u6309\u9a71\u52a8\u9009\uff09\uff1apip install torch torchvision torchaudio &#8211;index-url https:\/\/download.pytorch.org\/whl\/cu128<br \/>\n4. \u9a8c\u8bc1\uff1aimport torch; print(torch.cuda.is_available())\u00a0 \u8fd4\u56de True \u4e14\u975e +cpu \u5373\u5bf9\u3002<\/p>\n<p>4060 \/ 8G \u663e\u5b58\u4e13\u9879\uff1a\u8dd1 Flux \u8fd9\u7c7b\u5927\u6a21\u578b\uff0c\u542f\u52a8\u52a0&#8211;lowvram &#8211;reserve-vram 1.0\u00a0\uff08\u5355\u4f4d GB\uff0c\u7ed9\u7cfb\u7edf\u9884\u7559\u9632\u5d29\uff09\uff1b\u91cf\u5316\u9009\u8f7b\u91cf\u6863\uff0c4060 \u5b9e\u6d4b Q3_K_M \u7ea6 7.2G\u3001Q4_K_M \u7ea6 8.5G \u5df2\u8fd1\u6781\u9650\uff0c\u4f18\u5148\u66f4\u8f7b\u7684\u3002<\/p>\n<p><strong>\u901a\u7528\u81ea\u6551\u4e09\u6b65\u6cd5<\/strong><\/p>\n<p>1.\u00a0<strong>\u770b\u7ea2\u5b57<\/strong>\uff1a\u622a\u56fe\uff0c\u627e\u6700\u540e\u4e00\u884c\u5173\u952e\u8bcd\uff08CUDA \/ ModuleNotFound \/ Model not found \/ path\uff09\u3002<br \/>\n2.\u00a0<strong>\u641c\u5173\u952e\u8bcd<\/strong>\uff1a\u5173\u952e\u8bcd + &#8220;ComfyUI&#8221; \u641c\u4e00\u4e0b\uff0c\u57fa\u672c\u90fd\u6709\u4eba\u9047\u8fc7\u3002<br \/>\n3.\u00a0<strong>\u5bf9\u75c7\u5904\u7406<\/strong>\uff1a\u964d\u663e\u5b58 \/ \u88c5\u8282\u70b9 \/ \u653e\u5bf9\u76ee\u5f55 \/ \u91cd\u88c5 torch \/ \u6362 VAE \u964d CFG \/ \u6e05\u961f\u5217\u91cd\u542f\u3002<\/p>\n<p>\u539f\u5219\u5c31\u4e00\u53e5\uff1a<strong>\u6539\u4e00\u9879\u3001\u91cd\u542f\u4e00\u6b21\u3001\u56de\u6d4b\u4e00\u5f20\u56fe\u3002<\/strong>\u522b\u614c\uff0c\u7ea2\u5b57\u662f\u5e2e\u4f60\u5b9a\u4f4d\u7684\uff0c\u4e0d\u662f\u5413\u4f60\u7684\u3002<\/p>\n<p>\u62a5\u9519\u80fd\u81ea\u6551\u4e86\uff0c\u57fa\u7840\u4e5f\u624e\u5b9e\u4e86\u3002\u63a5\u4e0b\u6765\u73a9\u70b9\u66f4\u53ef\u63a7\u7684\u2014\u2014<strong>\u7b2c08\u7bc7<\/strong>\u8bb2<strong>\u56fe\u751f\u56fe\uff08img2img\uff09<\/strong>\uff1a\u62ff\u4e00\u5f20\u73b0\u6709\u56fe\u5f53\u5e95\u5b50\uff0c\u6539\u98ce\u683c\u3001\u6362\u6784\u56fe\u3001\u51fa\u53d8\u4f53\uff0c\u6bd4\u4ece\u96f6\u778e\u731c\u63d0\u793a\u8bcd\u9ad8\u6548\u5f97\u591a\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Previous (Page 06) ComfyUI, from installation entry to mastery, produced maps for quality improvement (resolution\/step\/sampler\/CFG\/high-resolution magnification) Parameters are adjusted and the paints are up. But playing AI drawings, who hasn't seen all the red words? When I first got wrong, I went straight down -- a lot of English was brushed down, it looked like the program went down and almost took the software off the reload. Then it turns out that ComfyUI's error is actually a \"promptology book\" and that most of the problems can be solved by reading the last line of the red word. This is one of the red words most often encountered by the newcomers: how can the phenomenon be repaired? Each of them takes me through the pit. It's recommended that the collection hit a direct flip one day. Wrong 1: Node type<\/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":[149,144],"tags":[1989,6276,4749],"collection":[],"class_list":["post-55758","post","type-post","status-publish","format-standard","hentry","category-jiaocheng","category-baike","tag-comfyui"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/55758","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=55758"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/55758\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=55758"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=55758"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=55758"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=55758"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}