{"id":104935,"date":"2026-09-13T13:25:23","date_gmt":"2026-09-13T05:25:23","guid":{"rendered":"https:\/\/www.wsisp.com\/helps\/104935.html"},"modified":"2026-09-13T13:25:23","modified_gmt":"2026-09-13T05:25:23","slug":"comfyui-v0-34-0-%e5%ae%8c%e5%85%a8%e9%83%a8%e7%bd%b2%e4%b8%8e3d%e7%94%9f%e6%88%90%e5%ae%9e%e6%88%98%e6%8c%87%e5%8d%97%ef%bc%9ahdr%e8%a7%86%e9%a2%91%e3%80%81trellis2%e5%bb%ba%e6%a8%a1%e4%b8%8eminimax-h","status":"publish","type":"post","link":"https:\/\/www.wsisp.com\/helps\/104935.html","title":{"rendered":"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357\uff1aHDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41"},"content":{"rendered":"

ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357：HDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41<\/h2>\n

\u4f5c\u8005：\u8d5b\u535a\u4ed3\u9f20<\/p>\n

\u4e00\u53e5\u8bdd\u603b\u7ed3：ComfyUI v0.34.0\u4e0d\u662f\u4e00\u6b21\u5c0f\u4fee\u5c0f\u8865\u2014\u2014\u5b83\u628aHDR\u89c6\u9891\u5bfc\u51fa\u30013D\u8d44\u4ea7\u751f\u6210\u3001\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u8fd9\u4e09\u4e2a\u539f\u672c\u9700\u8981N\u4e2a\u8f6f\u4ef6\u624d\u80fd\u5b8c\u6210\u7684\u6d41\u7a0b，\u5168\u90e8\u585e\u8fdb\u4e86\u4e00\u4e2a\u8282\u70b9\u56fe\u91cc\u3002\u5982\u679c\u4f60\u8fd8\u5728\u7528v0.33.x，\u8fd9\u7bc7\u6307\u5357\u5c31\u662f\u5347\u7ea7\u7684\u7406\u7531\u3002<\/p>\n

\n

\u4e00\u3001\u4e3a\u4ec0\u4e48\u4f60\u73b0\u5728\u5fc5\u987b\u5347\u7ea7\u5230v0.34.0<\/h3>\n

2026\u5e749\u67089\u65e5，ComfyUI\u53d1\u5e03\u4e86v0.34.0\u4e0d\u53ef\u53d8\u7248\u672c\u3002\u4f5c\u4e3aAI\u7ed8\u753b\u4e0e\u89c6\u9891\u751f\u6210\u9886\u57df\u7684\u4e8b\u5b9e\u6807\u51c6\u5de5\u4f5c\u6d41\u5e73\u53f0，\u8fd9\u6b21\u66f4\u65b0\u8986\u76d6\u4e86\u516d\u4e2a\u76f4\u63a5\u5f71\u54cd\u751f\u4ea7\u529b\u7684\u65b9\u5411：<\/p>\n

  • 3D\u8d44\u4ea7\u751f\u6210：\u539f\u751f\u652f\u6301TRELLIS2（\u5fae\u8f6f）\u548cPixal3d（\u6e05\u534e\/\u817e\u8bafARC），\u5355\u5f20\u56fe\u7247\u2192\u5e26PBR\u6750\u8d28\u76843D\u6a21\u578b<\/li>\n
  • HDR\u89c6\u9891\u5bfc\u51fa：\u65b0\u589eAV1\u7f16\u7801\u5668，\u652f\u6301HDR\u8272\u5f69\u7a7a\u95f4\u3001mkv\/webm\u5bb9\u5668，\u89c6\u9891\u753b\u8d28\u76f4\u63a5\u5bf9\u6807\u5f71\u89c6\u5de5\u4e1a\u6807\u51c6<\/li>\n
  • \u97f3\u89c6\u9891\u4e00\u4f53\u5316：MiniMax-H3\u652f\u6301\u9010Token\u89c6\u9891+\u97f3\u9891\u8054\u5408\u751f\u6210，\u753b\u9762\u548c\u97f3\u8f68\u5728\u6f5c\u7a7a\u95f4\u5c31\u88ab\u9501\u5b9a<\/li>\n
  • \u6a21\u578b\u751f\u6001\u6269\u5c55：\u65b0\u589eWan 3.0\u3001Seedance 2.5\u3001Seedream\u3001Pixverse V6\u7b49\u89c6\u9891\u6a21\u578b\u8282\u70b9<\/li>\n
  • Apple Silicon\u4f18\u5316：MLX\u8fd0\u884c\u80fd\u529b\u589e\u5f3a，\u7ed3\u6784\u5316\u8f93\u51fa\u6027\u80fd\u63d0\u5347，Mac\u7528\u6237\u7ec8\u4e8e\u4e0d\u7528\u7fa1\u6155N\u5361<\/li>\n
  • OpenAI\u517c\u5bb9：\u5de5\u5177\u641c\u7d22\u652f\u6301\u3001\u54cd\u5e94\u538b\u7f29\u3001\u4e0a\u4e0b\u6587\u538b\u7f29\u4f18\u5316\u2014\u2014\u957f\u4f1a\u8bdd\u56fe\u7247\u4e0d\u518d\u6d88\u5931<\/li>\n

    \u672c\u6587\u8986\u76d6\u4ece\u5b89\u88c5\u5347\u7ea7\u52303D\u751f\u6210\u5b9e\u6218\u7684\u5b8c\u6574\u94fe\u8def，\u6240\u6709\u4ee3\u7801\u548c\u5de5\u4f5c\u6d41\u5747\u53ef\u76f4\u63a5\u590d\u5236\u8fd0\u884c\u3002<\/p>\n

    \n

    \u4e8c\u3001\u73af\u5883\u51c6\u5907：\u5347\u7ea7\u524d\u5fc5\u505a\u7684\u4e09\u4ef6\u4e8b<\/h3>\n

    2.1 \u68c0\u67e5Python\u7248\u672c<\/h4>\n

    v0.34.0\u5b98\u65b9\u63d0\u793a：Python 3.10\u5373\u5c06EOL，\u5f3a\u70c8\u5efa\u8bae\u5347\u7ea7\u5230Python 3.11\u62163.12\u3002<\/p>\n

    # \u68c0\u67e5\u5f53\u524dPython\u7248\u672c<\/span>
    \npython –version<\/span>
    \n# \u5982\u679c\u663e\u793a Python 3.10.x，\u5efa\u8bae\u5347\u7ea7<\/span><\/p>\n

    # \u4f7f\u7528conda\u521b\u5efa\u65b0\u73af\u5883（\u63a8\u8350）<\/span>
    \nconda create -n<\/span> comfyui python<\/span>=<\/span>3.11<\/span>
    \nconda activate comfyui<\/p>\n

    # \u6216\u76f4\u63a5\u4f7f\u7528python.org\u5b89\u88c5\u5305\u8986\u76d6\u5347\u7ea7<\/span>
    \n# https:\/\/www.python.org\/downloads\/release\/python-3119\/<\/span><\/p>\n

    \u907f\u5751：Windows\u7528\u6237\u5347\u7ea7Python\u540e，\u9700\u8981\u91cd\u65b0\u5b89\u88c5torch\u548cCUDA\u4f9d\u8d56，\u4e0d\u8981\u76f4\u63a5\u590d\u7528\u65e7\u73af\u5883\u7684site-packages\u3002<\/p>\n

    2.2 \u5347\u7ea7ComfyUI\u672c\u4f53<\/h4>\n

    \u65b9\u6848A：Git\u76f4\u63a5\u5347\u7ea7（\u63a8\u8350）<\/p>\n

    # \u8fdb\u5165ComfyUI\u76ee\u5f55<\/span>
    \ncd<\/span> ComfyUI<\/p>\n

    # \u62c9\u53d6\u6700\u65b0\u4ee3\u7801<\/span>
    \ngit<\/span> pull origin master<\/p>\n

    # \u66f4\u65b0\u4f9d\u8d56<\/span>
    \npip install<\/span> -r<\/span> requirements.txt –upgrade<\/span><\/p>\n

    \u65b9\u6848B：\u79cb\u53f6\u6574\u5408\u5305\u7528\u6237<\/p>\n

    \u6253\u5f00\u6574\u5408\u5305\u81ea\u5e26\u7684"\u4e00\u952e\u66f4\u65b0"\u811a\u672c，\u6216\u624b\u52a8\u66ff\u6362\u4ee5\u4e0b\u76ee\u5f55：<\/p>\n

      \n
    • ComfyUI\/ \u4e3b\u7a0b\u5e8f\u76ee\u5f55<\/li>\n
    • ComfyUI\/custom_nodes\/ \u4e2d\u7684Manager\u63d2\u4ef6<\/li>\n<\/ul>\n

      \u907f\u5751：v0.34.0\u66f4\u65b0\u4e86\u524d\u7aef\u5305\u81f31.49.6，\u5982\u679c\u5347\u7ea7\u540e\u754c\u9762\u663e\u793a\u5f02\u5e38，\u5f3a\u5236\u5237\u65b0\u6d4f\u89c8\u5668\u7f13\u5b58（Ctrl+F5）\u6216\u6e05\u9664ComfyUI\/web\/assets\/\u76ee\u5f55\u540e\u91cd\u542f\u3002<\/p>\n

      2.3 \u663e\u5b58\u4e0e\u786c\u4ef6\u57fa\u51c6<\/h4>\n\n\u529f\u80fd\u6a21\u5757\u6700\u4f4e\u663e\u5b58\u63a8\u8350\u663e\u5b58\u6d4b\u8bd5\u5361\u578b<\/tr>\n\n\n\n\n\n\n\n
      \u6807\u51c6SDXL\u51fa\u56fe<\/td>\n8GB<\/td>\n12GB<\/td>\nRTX 4060<\/td>\n<\/tr>\n
      TRELLIS2 3D\u751f\u6210<\/td>\n24GB<\/td>\n24GB+<\/td>\nRTX 4090 \/ A6000<\/td>\n<\/tr>\n
      Pixal3d 3D\u751f\u6210<\/td>\n16GB<\/td>\n24GB<\/td>\nRTX 3090 \/ 4090<\/td>\n<\/tr>\n
      MiniMax-H3 \u89c6\u9891\u751f\u6210<\/td>\n12GB<\/td>\n16GB<\/td>\nRTX 4070 Ti<\/td>\n<\/tr>\n
      Wan 3.0 \u89c6\u9891\u751f\u6210<\/td>\n12GB<\/td>\n16GB<\/td>\nRTX 4070 Ti<\/td>\n<\/tr>\n
      HDR\u89c6\u9891\u5bfc\u51fa<\/td>\n8GB<\/td>\n12GB<\/td>\n\u4efb\u610fN\u5361<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

      \u6ce8\u610f：TRELLIS2\u5b98\u65b9\u8981\u6c4224GB\u663e\u5b58，\u4f46ComfyUI v0.34.0\u7684\u8282\u70b9\u5b9e\u73b0\u652f\u6301\u5185\u5b58\u56de\u9000（memory offloading），16GB\u5361\u53ef\u4ee5\u8dd1，\u901f\u5ea6\u61623-4\u500d\u3002<\/p>\n

      \n

      \u4e09\u30013D\u8d44\u4ea7\u751f\u6210\u5b9e\u6218：TRELLIS2 + Pixal3d\u53cc\u65b9\u6848<\/h3>\n

      3.1 TRELLIS2：\u5fae\u8f6f\u76843D\u8d44\u4ea7\u751f\u6210\u5668<\/h4>\n

      \u6280\u672f\u80cc\u666f：TRELLIS.2\u7531\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u4e8e2025\u5e7412\u6708\u5f00\u6e90，\u91c7\u7528SLAT（Structured Latent Attention Transformers）\u8868\u793a\u6cd5，\u6838\u5fc3\u4f18\u52bf\u662f\u4ece\u5355\u5f20\u56fe\u7247\u751f\u6210\u5e26PBR\u6750\u8d28\u7684\u9ad8\u8d28\u91cf3D\u8d44\u4ea7\u3002<\/p>\n

      \u652f\u6301\u683c\u5f0f：<\/p>\n

        \n
      • \u8f90\u5c04\u573a（Radiance Field）<\/li>\n
      • 3D\u9ad8\u65af（3D Gaussian Splatting）<\/li>\n
      • \u7f51\u683c（Mesh）\u2192 \u5bfc\u51faGLB\u3001PLY<\/li>\n<\/ul>\n

        ComfyUI v0.34.0\u539f\u751f\u8282\u70b9：<\/p>\n

          \n
        • TRELLIS2 Loader：\u52a0\u8f7dTRELLIS.2-4B\u6a21\u578b<\/li>\n
        • TRELLIS2 Image to 3D：\u56fe\u7247\u8f93\u5165\u21923D\u8d44\u4ea7\u8f93\u51fa<\/li>\n
        • TRELLIS2 Material Export：\u5bfc\u51faPBR\u6750\u8d28\u8d34\u56fe<\/li>\n<\/ul>\n

          \u5b89\u88c5\u6b65\u9aa4：<\/p>\n

          # 1. \u4e0b\u8f7dTRELLIS.2-4B\u9884\u8bad\u7ec3\u6a21\u578b<\/span>
          \n# \u6a21\u578b\u5730\u5740：https:\/\/huggingface.co\/microsoft\/TRELLIS.2-4B<\/span>
          \n# \u6587\u4ef6\u5927\u5c0f：\u7ea618GB<\/span><\/p>\n

          # 2. \u653e\u7f6e\u5230ComfyUI\u6a21\u578b\u76ee\u5f55<\/span>
          \nmkdir<\/span> -p<\/span> ComfyUI\/models\/trellis2
          \ncp<\/span> TRELLIS.2-4B\/* ComfyUI\/models\/trellis2\/<\/p>\n

          # 3. \u5b89\u88c5\u4f9d\u8d56<\/span>
          \npip install<\/span> kaolin pytorch3d trimesh
          \n# \u6ce8\u610f：kaolin\u5b89\u88c5\u8f83\u590d\u6742，\u5982\u679c\u7f16\u8bd1\u5931\u8d25，\u4f7f\u7528\u9884\u7f16\u8bd1wheel<\/span>
          \n# pip install kaolin -f https:\/\/nvidia-kaolin.s3.us-east-2.amazonaws.com\/torch-2.4.0_cu121.html<\/span><\/p>\n

          \u6700\u5c0f\u5de5\u4f5c\u6d41\u642d\u5efa：<\/p>\n

          [Load Image] \u2192 [TRELLIS2 Image to 3D] \u2192 [TRELLIS2 Material Export] \u2192 [Save 3D Asset]
          \n \u2191
          \n [TRELLIS2 Loader]
          \n (model: TRELLIS.2-4B)<\/p>\n

          \u8282\u70b9\u53c2\u6570\u8bf4\u660e：<\/p>\n\n\u53c2\u6570\u9ed8\u8ba4\u503c\u8bf4\u660e<\/tr>\n\n\n\n\n\n\n
          seed<\/td>\n42<\/td>\n\u968f\u673a\u79cd\u5b50，\u5f71\u54cd3D\u7ec6\u8282\u5206\u5e03<\/td>\n<\/tr>\n
          steps<\/td>\n50<\/td>\n\u91c7\u6837\u6b65\u6570，30-50\u4e4b\u95f4\u8d28\u91cf\/\u901f\u5ea6\u5e73\u8861<\/td>\n<\/tr>\n
          cfg_scale<\/td>\n7.5<\/td>\n\u5206\u7c7b\u5668\u81ea\u7531\u5f15\u5bfc\u5c3a\u5ea6<\/td>\n<\/tr>\n
          output_format<\/td>\n\u201cglb\u201d<\/td>\n\u8f93\u51fa\u683c\u5f0f：glb \/ ply \/ gaussian<\/td>\n<\/tr>\n
          material_quality<\/td>\n\u201chigh\u201d<\/td>\nPBR\u6750\u8d28\u8d28\u91cf：low\/medium\/high<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

          \u5b9e\u6d4b\u6570\u636e（RTX 4090 24GB）：<\/p>\n

            \n
          • \u5355\u5f201024\u00d71024\u56fe\u7247 \u2192 3D GLB\u8d44\u4ea7：\u7ea645\u79d2<\/li>\n
          • \u751f\u6210\u6587\u4ef6\u5927\u5c0f：GLB\u7ea615-30MB（\u542bPBR\u8d34\u56fe）<\/li>\n
          • \u663e\u5b58\u5cf0\u503c\u5360\u7528：\u7ea622GB<\/li>\n<\/ul>\n

            3.2 Pixal3d：\u50cf\u7d20\u5bf9\u9f50\u76843D\u91cd\u5efa<\/h4>\n

            \u6280\u672f\u80cc\u666f：Pixal3d\u7531\u6e05\u534e\u5927\u5b66\u4e0e\u817e\u8bafARC\u8054\u5408\u7814\u53d1，SIGGRAPH 2026\u6536\u5f55\u3002\u6838\u5fc3\u521b\u65b0\u662f"Pixel-Aligned"（\u50cf\u7d20\u5bf9\u9f50）\u2014\u2014\u8f93\u51fa3D\u6a21\u578b\u5728\u50cf\u7d20\u7ea7\u522b\u4e0e\u8f93\u5165\u56fe\u7247\u5bf9\u9f50，\u4e0d\u4f1a\u51fa\u73b0\u4f20\u7edf3D\u751f\u6210"\u6b63\u9762\u50cf\u3001\u80cc\u9762\u5d29"\u7684\u95ee\u9898\u3002<\/p>\n

            \u4e0eTRELLIS2\u7684\u6838\u5fc3\u5dee\u5f02：<\/p>\n\n\u7ef4\u5ea6TRELLIS2Pixal3d<\/tr>\n\n\n\n\n\n\n\n
            \u67b6\u6784\u57fa\u7840<\/td>\nSLAT<\/td>\nTrellis.2\u4f18\u5316\u7248<\/td>\n<\/tr>\n
            \u50cf\u7d20\u5bf9\u9f50<\/td>\n\u4e00\u822c<\/td>\n\u5f3a（\u6838\u5fc3\u5356\u70b9）<\/td>\n<\/tr>\n
            \u573a\u666f\u590d\u6742\u5ea6<\/td>\n\u5355\u4f53\u4e3a\u4e3b<\/td>\n\u652f\u6301\u591a\u7269\u4f53\u573a\u666f<\/td>\n<\/tr>\n
            \u6750\u8d28\u7cfb\u7edf<\/td>\nPBR\u5b8c\u6574<\/td>\nPBR\u5b8c\u6574<\/td>\n<\/tr>\n
            \u663e\u5b58\u9700\u6c42<\/td>\n24GB<\/td>\n16GB\u8d77<\/td>\n<\/tr>\n
            \u751f\u6210\u901f\u5ea6<\/td>\n45s\/\u6a21\u578b<\/td>\n30s\/\u6a21\u578b<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

            ComfyUI\u8282\u70b9\u5b89\u88c5：<\/p>\n

            Pixal3d\u76ee\u524d\u4ee5\u793e\u533a\u8282\u70b9\u5f62\u5f0f\u63d0\u4f9b，\u9700\u624b\u52a8\u5b89\u88c5：<\/p>\n

            cd<\/span> ComfyUI\/custom_nodes
            \ngit<\/span> clone https:\/\/github.com\/pixal3d\/ComfyUI-Pixal3d.git<\/p>\n

            # \u5b89\u88c5\u4f9d\u8d56<\/span>
            \ncd<\/span> ComfyUI-Pixal3d
            \npip install<\/span> -r<\/span> requirements.txt<\/p>\n

            # \u4e0b\u8f7d\u6a21\u578b\u6743\u91cd（\u7ea612GB）<\/span>
            \n# https:\/\/huggingface.co\/pixal3d\/pixal3d-v1<\/span>
            \nmkdir<\/span> -p<\/span> ..<\/span>\/..<\/span>\/models\/pixal3d
            \ncp<\/span> pixal3d-v1\/* ..<\/span>\/..<\/span>\/models\/pixal3d\/<\/p>\n

            \u591a\u7269\u4f53\u573a\u666f\u5de5\u4f5c\u6d41：<\/p>\n

            [Load Image] \u2192 [Pixal3d Scene Parser] \u2192 [Pixal3d Object Separator]
            \n \u2193
            \n [Pixal3d Multi-Object Generator] \u2192 [Scene Composer] \u2192 [Save 3D Scene]
            \n \u2191
            \n [Pixal3d Loader]<\/p>\n

            \u5173\u952e\u53c2\u6570：<\/p>\n

              \n
            • scene_complexity：\u201csimple\u201d（\u5355\u7269\u4f53）\/ \u201cmoderate\u201d（2-3\u7269\u4f53）\/ \u201ccomplex\u201d（4+\u7269\u4f53）<\/li>\n
            • alignment_strictness：0.0-1.0，\u8d8a\u9ad8\u50cf\u7d20\u5bf9\u9f50\u8d8a\u4e25\u683c，\u4f46\u53ef\u80fd\u4e22\u5931\u906e\u6321\u533a\u57df\u7ec6\u8282<\/li>\n<\/ul>\n

              3.3 3D\u8d44\u4ea7\u751f\u6210\u907f\u5751\u6307\u5357<\/h4>\n

              \u5751\u70b91：kaolin\u5b89\u88c5\u5931\u8d25<\/p>\n

                \n
              • \u75c7\u72b6：pip install kaolin\u7f16\u8bd1\u62a5\u9519<\/li>\n
              • \u89e3\u51b3：\u4f7f\u7528NVIDIA\u9884\u7f16\u8bd1wheel，\u6216\u6539\u7528Docker\u73af\u5883<\/li>\n
              • \u5907\u7528：\u5982\u679c\u672c\u5730\u8dd1\u4e0d\u52a8，\u76f4\u63a5\u7528TRELLIS2\u7684Hugging Face Demo：https:\/\/huggingface.co\/spaces\/microsoft\/TRELLIS.2<\/li>\n<\/ul>\n

                \u5751\u70b92：GLB\u5bfc\u5165Blender\u540e\u6750\u8d28\u4e22\u5931<\/p>\n

                  \n
                • \u539f\u56e0：ComfyUI\u5bfc\u51fa\u7684PBR\u8d34\u56fe\u8def\u5f84\u4f7f\u7528\u4e86\u7edd\u5bf9\u8def\u5f84<\/li>\n
                • \u89e3\u51b3：\u5bfc\u51fa\u65f6\u52fe\u9009"Embed Textures"（\u5185\u5d4c\u8d34\u56fe），\u6216\u624b\u52a8\u5c06textures\u6587\u4ef6\u5939\u4e0eGLB\u653e\u5728\u540c\u7ea7\u76ee\u5f55<\/li>\n<\/ul>\n

                  \u5751\u70b93：16GB\u5361\u8dd1TRELLIS2\u7206\u663e\u5b58<\/p>\n

                    \n
                  • \u89e3\u51b3：\u5728ComfyUI\u542f\u52a8\u53c2\u6570\u4e2d\u6dfb\u52a0 –lowvram<\/li>\n
                  • \u6216\u6539\u7528Pixal3d（16GB\u53ef\u8dd1）<\/li>\n<\/ul>\n
                    \n

                    \u56db\u3001HDR\u89c6\u9891\u4e0e\u97f3\u89c6\u9891\u4e00\u4f53\u5316：MiniMax-H3 + Wan 3.0\u5b9e\u6218<\/h3>\n

                    4.1 HDR\u89c6\u9891\u5bfc\u51fa：\u753b\u8d28\u8dc3\u5347\u7684\u5173\u952e\u5347\u7ea7<\/h4>\n

                    v0.34.0\u4e4b\u524d，ComfyUI\u4fdd\u5b58\u7684\u89c6\u9891\u90fd\u662fSDR（\u6807\u51c6\u52a8\u6001\u8303\u56f4），\u9ad8\u5149\u7ec6\u8282\u5bb9\u6613\u8fc7\u66dd\u3002v0.34.0\u65b0\u589e：<\/p>\n

                      \n
                    • AV1\u7f16\u7801\u5668：\u6bd4H.264\u8282\u770140%\u7801\u7387，\u540c\u7b49\u753b\u8d28\u6587\u4ef6\u66f4\u5c0f<\/li>\n
                    • HDR\u8272\u5f69\u7a7a\u95f4：\u652f\u6301HLG\u548cPQ\u66f2\u7ebf<\/li>\n
                    • \u5bb9\u5668\u683c\u5f0f\u6269\u5c55：mkv\u3001webm（\u4e4b\u524d\u53ea\u6709mp4）<\/li>\n
                    • bit_depth\u4e0b\u62c9\u9009\u62e9：8bit\/10bit\/12bit\u53ef\u9009<\/li>\n<\/ul>\n

                      \u5de5\u4f5c\u6d41\u8282\u70b9\u914d\u7f6e：<\/p>\n

                      [Video Generation Pipeline] \u2192 [Save Video]
                      \n \u2193
                      \n codec: "av1"
                      \n colorspace: "HDR10"
                      \n bit_depth: "10bit"
                      \n container: "mkv"<\/p>\n

                      \u786c\u4ef6\u8981\u6c42：<\/p>\n

                        \n
                      • \u7f16\u7801：RTX 40\u7cfb\u5217\u652f\u6301AV1\u786c\u4ef6\u7f16\u7801，\u901f\u5ea6\u63d0\u53473\u500d<\/li>\n
                      • \u89e3\u7801\u64ad\u653e：\u9700\u8981\u652f\u6301HDR\u7684\u663e\u793a\u5668+\u64ad\u653e\u5668（PotPlayer\/VLC\u9700\u5f00\u542fHDR\u6a21\u5f0f）<\/li>\n<\/ul>\n

                        4.2 MiniMax-H3：\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u751f\u6210<\/h4>\n

                        MiniMax-H3\u7684\u6838\u5fc3\u521b\u65b0\u662f\u628a\u89c6\u9891\u548c\u97f3\u9891\u653e\u5728\u540c\u4e00\u4e2a\u6f5c\u7a7a\u95f4\u8054\u5408\u5efa\u6a21\u3002\u4f20\u7edf\u8def\u7ebf\u662f"\u5148\u751f\u6210\u753b\u9762\u2192\u518d\u7528\u5176\u4ed6\u5de5\u5177\u914d\u97f3\u2192\u624b\u52a8\u5bf9\u9f50"，H3\u5219\u662f\u5728\u751f\u6210\u9636\u6bb5\u5c31\u628a\u97f3\u8f68\u548c\u89c6\u8f68\u9501\u5b9a\u5728\u540c\u4e00\u4e2a\u65f6\u95f4\u8f74\u4e0a\u3002<\/p>\n

                        ComfyUI v0.34.0\u65b0\u589eH3\u8282\u70b9：<\/p>\n

                          \n
                        • MiniMaxH3 Loader：\u52a0\u8f7dH3\u4e3b\u6a21\u578b\u548cVAE<\/li>\n
                        • MiniMaxH3 Text to Video Audio：\u6587\u751f\u89c6\u9891+\u97f3\u9891<\/li>\n
                        • MiniMaxH3 Image to Video Audio：\u56fe\u751f\u89c6\u9891+\u97f3\u9891<\/li>\n
                        • MiniMaxH3 Audio Guide：\u97f3\u9891\u5f15\u5bfc\u8282\u70b9（\u63a7\u5236\u751f\u6210\u98ce\u683c）<\/li>\n
                        • MiniMaxH3 Noise Mask：\u9010Token\u89c6\u9891\/\u97f3\u9891\u6f5c\u53d8\u91cf\u566a\u58f0\u906e\u7f69<\/li>\n<\/ul>\n

                          \u6700\u5c0f\u95ed\u73af\u5de5\u4f5c\u6d41：<\/p>\n

                          [Prompt Text] \u2192 [MiniMaxH3 Text to Video Audio] \u2192 [Video Preview]
                          \n \u2191
                          \n [MiniMaxH3 Loader]
                          \n [VAE Loader]<\/p>\n

                          \u5173\u952e\u53c2\u6570\u8c03\u4f18：<\/p>\n\n\u53c2\u6570\u9ed8\u8ba4\u503c\u8c03\u4f18\u5efa\u8bae<\/tr>\n\n\n\n\n\n\n
                          frame_rate<\/td>\n25<\/td>\n\u5efa\u8bae25\u621630，\u5fc5\u987b\u4e0e\u97f3\u9891\u91c7\u6837\u7387\u5bf9\u9f50<\/td>\n<\/tr>\n
                          audio_sample_rate<\/td>\n44100<\/td>\n\u53ef\u900944100\u621648000，\u4e0e\u89c6\u9891\u5e27\u7387\u5339\u914d<\/td>\n<\/tr>\n
                          video_steps<\/td>\n30<\/td>\n25-40\u4e4b\u95f4，\u8d8a\u9ad8\u753b\u8d28\u8d8a\u7a33<\/td>\n<\/tr>\n
                          audio_guidance_scale<\/td>\n3.0<\/td>\n1.0-5.0，\u63a7\u5236\u97f3\u9891\u5bf9\u89c6\u9891\u7684\u5f71\u54cd\u529b<\/td>\n<\/tr>\n
                          sync_strength<\/td>\n0.8<\/td>\n0.0-1.0，\u97f3\u89c6\u9891\u540c\u6b65\u5f3a\u5ea6<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                          \u97f3\u89c6\u9891\u540c\u6b65\u907f\u5751：<\/p>\n

                          \u5982\u679c\u4f60\u9047\u5230"\u753b\u9762\u5df2\u7ed3\u675f，\u97f3\u9891\u8fd8\u5728\u64ad"\u7684\u60c5\u51b5，\u68c0\u67e5：<\/p>\n

                        • frame_rate\u5fc5\u987b\u662f25\u621630（\u4e0d\u80fd\u752824\u621660）<\/li>\n
                        • audio_sample_rate\u5fc5\u987b\u662f44100Hz\u621648000Hz<\/li>\n
                        • \u89c6\u9891\u603b\u5e27\u6570 = \u5e27\u7387 \u00d7 \u65f6\u957f（\u79d2），\u97f3\u9891\u6837\u672c\u6570 = \u91c7\u6837\u7387 \u00d7 \u65f6\u957f（\u79d2），\u4e24\u8005\u65f6\u957f\u5fc5\u987b\u4e25\u683c\u4e00\u81f4<\/li>\n

                          4.3 Wan 3.0：\u5f00\u6e90\u89c6\u9891\u751f\u6210\u65b0\u6807\u6746<\/h4>\n

                          Wan 3.0\u662f\u963f\u91cc\u901a\u4e49\u4e07\u76f8\u56e2\u961f\u5f00\u6e90\u7684\u89c6\u9891\u751f\u6210\u6a21\u578b，v0.34.0\u539f\u751f\u652f\u6301：<\/p>\n

                          [Wan3 Loader] \u2192 [Wan3 Text to Video] \u2192 [Save Video]<\/p>\n

                          \u6027\u80fd\u5bf9\u6bd4（RTX 4090，720p，81\u5e27）：<\/p>\n\n\u6a21\u578b\u751f\u6210\u65f6\u95f4\u663e\u5b58\u5360\u7528\u753b\u8d28\u8bc4\u5206<\/tr>\n\n\n\n\n\n
                          Wan 3.0 T2V 14B<\/td>\n8\u5206\u949f<\/td>\n20GB<\/td>\n\u2605\u2605\u2605\u2605\u2605<\/td>\n<\/tr>\n
                          Wan 3.0 T2V 1.3B<\/td>\n2\u5206\u949f<\/td>\n10GB<\/td>\n\u2605\u2605\u2605\u2605\u2606<\/td>\n<\/tr>\n
                          MiniMax-H3<\/td>\n5\u5206\u949f<\/td>\n14GB<\/td>\n\u2605\u2605\u2605\u2605\u2605<\/td>\n<\/tr>\n
                          Seedance 2.5<\/td>\n6\u5206\u949f<\/td>\n16GB<\/td>\n\u2605\u2605\u2605\u2605\u2606<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                          \u5efa\u8bae：16GB\u4ee5\u4e0b\u5361\u7528Wan 3.0 1.3B\u7248\u672c，24GB\u5361\u76f4\u63a5\u4e0a14B\u6216MiniMax-H3\u3002<\/p>\n

                          \n

                          \u4e94\u3001\u6027\u80fd\u4f18\u5316：\u8ba9ComfyUI\u5728\u6d88\u8d39\u7ea7\u663e\u5361\u4e0a\u98de\u8d77\u6765<\/h3>\n

                          5.1 \u52a8\u6001VRAM\u7ba1\u7406（v0.34.0\u9ed8\u8ba4\u5f00\u542f）<\/h4>\n

                          v0.34.0\u5bf9ROCm 7.14+\u9ed8\u8ba4\u542f\u7528\u52a8\u6001VRAM，N\u5361\u7528\u6237\u4e5f\u53ef\u624b\u52a8\u5f00\u542f：<\/p>\n

                          # \u542f\u52a8ComfyUI\u65f6\u6dfb\u52a0\u53c2\u6570<\/span>
                          \npython main.py –lowvram<\/span> –gpu-only<\/p>\n

                          # \u6216\u4fee\u6539\u914d\u7f6e\u6587\u4ef6 ComfyUI\/extra_model_paths.yaml<\/span>
                          \nvram_management: "dynamic"<\/span><\/p>\n

                          \u6548\u679c：12GB\u5361\u53ef\u4ee5\u8dd1\u539f\u672c\u9700\u898116GB\u7684\u5de5\u4f5c\u6d41，\u901f\u5ea6\u635f\u5931\u7ea615-20%\u3002<\/p>\n

                          5.2 \u4e0a\u4e0b\u6587\u538b\u7f29\u4f18\u5316（\u957f\u4f1a\u8bdd\u4e0d\u518d\u4e22\u56fe）<\/h4>\n

                          v0.34.0\u4fee\u590d\u4e86\u4e0a\u4e0b\u6587\u538b\u7f29\u65f6"\u56fe\u7247\u4ece\u56de\u590d\u91cc\u6d88\u5931"\u7684bug\u3002\u5982\u679c\u4f60\u7528ComfyUI\u505a\u591a\u8f6e\u5bf9\u8bdd\u5f0f\u751f\u6210（\u5982\u8fed\u4ee3\u4fee\u6539\u63d0\u793a\u8bcd），\u73b0\u5728\u56fe\u7247\u4f1a\u6b63\u786e\u4fdd\u7559\u3002<\/p>\n

                          \u542f\u7528\u65b9\u6cd5：\u65e0\u9700\u64cd\u4f5c，\u5347\u7ea7\u540e\u81ea\u52a8\u751f\u6548\u3002<\/p>\n

                          5.3 Windows\u591aGPU\u884c\u4e3a\u53d8\u5316<\/h4>\n

                          v0.34.0\u4fee\u6539\u4e86Windows\u4e0b\u591aGPU\u7684\u9ed8\u8ba4\u884c\u4e3a：<\/p>\n

                            \n
                          • \u65e7\u7248：\u81ea\u52a8\u5206\u914d\u4efb\u52a1\u5230\u6240\u6709\u53ef\u7528GPU<\/li>\n
                          • v0.34.0：\u9ed8\u8ba4\u53ea\u4f7f\u7528\u7b2c\u4e00\u5f20CUDA\u8bbe\u5907<\/li>\n<\/ul>\n

                            \u5982\u679c\u4f60\u6709\u591a\u5361（\u5982\u53ccRTX 3090），\u9700\u8981\u624b\u52a8\u6307\u5b9a：<\/p>\n

                            # \u4f7f\u7528\u7b2c0\u5f20\u548c\u7b2c1\u5f20\u5361<\/span>
                            \nset<\/span> CUDA_VISIBLE_DEVICES<\/span>=<\/span>0,1<\/span>
                            \npython main.py –multi-gpu<\/p>\n

                            # \u6216\u53ea\u4f7f\u7528\u7b2c1\u5f20\u5361（\u5982\u679c\u7b2c0\u5f20\u88ab\u5360\u7528）<\/span>
                            \nset<\/span> CUDA_VISIBLE_DEVICES<\/span>=<\/span>1<\/span>
                            \npython main.py<\/p>\n

                            \n

                            \u516d\u3001\u4ece0\u52301：\u642d\u5efa\u4f60\u7684\u7b2c\u4e00\u4e2a3D+\u89c6\u9891\u6df7\u5408\u5de5\u4f5c\u6d41<\/h3>\n

                            6.1 \u573a\u666f\u8bbe\u5b9a<\/h4>\n

                            \u5047\u8bbe\u4f60\u662f\u4e00\u4e2a\u6e38\u620f\u8d44\u4ea7\u8bbe\u8ba1\u5e08，\u9700\u8981\u4e3a\u4e00\u4e2a\u89d2\u8272\u521b\u5efa：<\/p>\n

                          • 3D\u89d2\u8272\u6a21\u578b（\u5e26PBR\u6750\u8d28）<\/li>\n
                          • \u89d2\u8272\u5c55\u793a\u89c6\u9891（360\u5ea6\u65cb\u8f6c）<\/li>\n
                          • \u89d2\u8272\u914d\u97f3\u5ba3\u4f20\u77ed\u7247<\/li>\n

                            6.2 \u5de5\u4f5c\u6d41\u642d\u5efa<\/h4>\n

                            \u9636\u6bb51：3D\u89d2\u8272\u751f\u6210<\/p>\n

                            [\u89d2\u8272\u7acb\u7ed8] \u2192 [TRELLIS2 Image to 3D] \u2192 [\u5bfc\u51faGLB]
                            \n \u2191
                            \n [TRELLIS2 Loader]<\/p>\n

                            \u9636\u6bb52：360\u5ea6\u5c55\u793a\u89c6\u9891<\/p>\n

                            \u5c06GLB\u5bfc\u5165Blender，\u6e32\u67d3360\u5ea6 turntable \u52a8\u753b（8\u79d2，30fps），\u5bfc\u51fa\u4e3aPNG\u5e8f\u5217\u3002<\/p>\n

                            \u9636\u6bb53：\u914d\u97f3\u5ba3\u4f20\u77ed\u7247<\/p>\n

                            [Turntable PNG\u5e8f\u5217] \u2192 [Load Image Batch] \u2192 [Wan3 Image to Video]
                            \n \u2193
                            \n [MiniMaxH3 Audio Guide]
                            \n \u2193
                            \n [Save Video (AV1\/HDR)]<\/p>\n

                            \u5b8c\u6574\u7ba1\u7ebf\u65f6\u95f4（RTX 4090）：<\/p>\n

                              \n
                            • 3D\u751f\u6210：45\u79d2<\/li>\n
                            • 360\u5ea6\u6e32\u67d3：5\u5206\u949f（Blender）<\/li>\n
                            • \u89c6\u9891+\u97f3\u9891\u751f\u6210：8\u5206\u949f<\/li>\n
                            • \u603b\u8ba1：\u7ea614\u5206\u949f\u5b8c\u6210\u4e00\u4e2a\u5e26\u914d\u97f3\u76843D\u89d2\u8272\u5c55\u793a\u77ed\u7247<\/li>\n<\/ul>\n
                              \n

                              \u4e03\u3001\u907f\u5751\u5927\u5168：v0.34.0\u5347\u7ea7\u5fc5\u770b<\/h3>\n

                              7.1 \u524d\u7aef\u663e\u793a\u5f02\u5e38<\/h4>\n

                              \u75c7\u72b6：\u5347\u7ea7\u540e\u754c\u9762\u7a7a\u767d\u6216\u8282\u70b9\u663e\u793a\u9519\u4e71
                              \n\u89e3\u51b3：<\/p>\n

                              # \u6e05\u9664\u524d\u7aef\u7f13\u5b58<\/span>
                              \ncd<\/span> ComfyUI\/web\/assets
                              \nrm<\/span> -rf<\/span> *<\/p>\n

                              # \u91cd\u65b0\u542f\u52a8ComfyUI，\u524d\u7aef\u4f1a\u81ea\u52a8\u91cd\u65b0\u7f16\u8bd1<\/span><\/p>\n

                              7.2 \u8282\u70b9\u7f3a\u5931\u62a5\u9519<\/h4>\n

                              \u75c7\u72b6：\u52a0\u8f7d\u5de5\u4f5c\u6d41\u65f6\u63d0\u793a"Node not found: MiniMaxH3\u2026"
                              \n\u89e3\u51b3：<\/p>\n

                              # 1. \u786e\u4fddManager\u63d2\u4ef6\u5df2\u66f4\u65b0<\/span>
                              \ncd<\/span> ComfyUI\/custom_nodes\/ComfyUI-Manager
                              \ngit<\/span> pull<\/p>\n

                              # 2. \u5728ComfyUI\u754c\u9762\u4e2d\u6253\u5f00Manager \u2192 Install Missing Custom Nodes<\/span>
                              \n# 3. \u91cd\u542fComfyUI<\/span><\/p>\n

                              7.3 \u6a21\u578b\u4e0b\u8f7d\u5931\u8d25<\/h4>\n

                              \u75c7\u72b6：Hugging Face\u6a21\u578b\u4e0b\u8f7d\u4e2d\u65ad
                              \n\u89e3\u51b3：<\/p>\n

                              # \u4f7f\u7528\u955c\u50cf\u7ad9<\/span>
                              \nexport<\/span> HF_ENDPOINT<\/span>=<\/span>https:\/\/hf-mirror.com
                              \npython -c<\/span> "from huggingface_hub import snapshot_download; snapshot_download('microsoft\/TRELLIS.2-4B')"<\/span><\/p>\n

                              # \u6216\u624b\u52a8\u4e0b\u8f7d\u540e\u653e\u7f6e\u5230\u5bf9\u5e94\u76ee\u5f55<\/span><\/p>\n

                              7.4 Apple Silicon\u7528\u6237\u7279\u522b\u8bf4\u660e<\/h4>\n

                              v0.34.0\u5bf9Apple Silicon\u7684\u7ed3\u6784\u5316\u8f93\u51fa\u6027\u80fd\u8fdb\u884c\u4e86\u4f18\u5316，\u4f463D\u751f\u6210（TRELLIS2\/Pixal3d）\u4ecd\u9700\u8981MLX\u8f6c\u6362\u540e\u7684\u6a21\u578b：<\/p>\n

                              # M\u7cfb\u5217Mac\u7528\u6237\u9700\u4e0b\u8f7dMLX\u7248\u672c\u6a21\u578b<\/span>
                              \n# https:\/\/huggingface.co\/microsoft\/TRELLIS.2-4B-MLX<\/span><\/p>\n

                              # \u663e\u5b58\u9650\u5236：M3 Max 36GB\u7edf\u4e00\u5185\u5b58\u53ef\u8dd1TRELLIS2（\u901f\u5ea6\u6bd44090\u616260%）<\/span>
                              \n# M3 Pro 18GB\u5efa\u8bae\u7528Pixal3d<\/span><\/p>\n

                              7.5 Python 3.10 EOL\u5f71\u54cd<\/h4>\n

                              \u867d\u7136v0.34.0\u4ecd\u652f\u6301Python 3.10，\u4f46\u540e\u7eed\u7248\u672c\u53ef\u80fd\u653e\u5f03\u652f\u6301\u3002\u5efa\u8bae：<\/p>\n

                              # \u5907\u4efd\u65e7\u73af\u5883<\/span>
                              \nconda env<\/span> export<\/span> ><\/span> comfyui_old.yml<\/p>\n

                              # \u521b\u5efa\u65b0\u73af\u5883<\/span>
                              \nconda create -n<\/span> comfyui311 python<\/span>=<\/span>3.11<\/span>
                              \nconda activate comfyui311<\/p>\n

                              # \u91cd\u65b0\u5b89\u88c5\u4f9d\u8d56<\/span>
                              \npip install<\/span> torch torchvision torchaudio –index-url https:\/\/download.pytorch.org\/whl\/cu121
                              \npip install<\/span> -r<\/span> ComfyUI\/requirements.txt<\/p>\n

                              \n

                              \u516b\u3001\u8d44\u6e90\u6c47\u603b<\/h3>\n

                              8.1 \u6a21\u578b\u4e0b\u8f7d\u5730\u5740<\/h4>\n\n\u6a21\u578b\u5730\u5740\u5927\u5c0f<\/tr>\n\n\n\n\n\n\n
                              TRELLIS.2-4B<\/td>\nhttps:\/\/huggingface.co\/microsoft\/TRELLIS.2-4B<\/td>\n18GB<\/td>\n<\/tr>\n
                              Pixal3d-v1<\/td>\nhttps:\/\/huggingface.co\/pixal3d\/pixal3d-v1<\/td>\n12GB<\/td>\n<\/tr>\n
                              MiniMax-H3<\/td>\nhttps:\/\/huggingface.co\/MiniMax\/MiniMax-H3<\/td>\n\u7ea615GB<\/td>\n<\/tr>\n
                              Wan 3.0 14B<\/td>\nhttps:\/\/huggingface.co\/Wan-AI\/Wan3.0-T2V-14B<\/td>\n28GB<\/td>\n<\/tr>\n
                              Wan 3.0 1.3B<\/td>\nhttps:\/\/huggingface.co\/Wan-AI\/Wan3.0-T2V-1.3B<\/td>\n3GB<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                              8.2 \u7f51\u76d8\u8d44\u6e90（\u5df2\u6574\u5408\u6240\u6709\u6a21\u578b+\u5de5\u4f5c\u6d41\u6a21\u677f）<\/h4>\n

                              \u5938\u514b\u7f51\u76d8：<\/p>\n

                                \n
                              • https:\/\/pan.quark.cn\/s\/a7d5cb0d9fbe<\/li>\n
                              • https:\/\/pan.quark.cn\/s\/a8a75ed5ef5a<\/li>\n<\/ul>\n

                                \u7f51\u76d8\u5185\u5305\u542b：<\/p>\n

                                  \n
                                • ComfyUI v0.34.0 Windows\u6574\u5408\u5305（\u9884\u88c5\u6240\u67093D\/\u89c6\u9891\u8282\u70b9）<\/li>\n
                                • TRELLIS2 + Pixal3d + MiniMax-H3 + Wan 3.0 \u6a21\u578b\u6587\u4ef6<\/li>\n
                                • 5\u4e2a\u53ef\u76f4\u63a5\u5bfc\u5165\u7684\u5de5\u4f5c\u6d41\u6a21\u677f（3D\u751f\u6210\u3001HDR\u89c6\u9891\u3001\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u3001\u6279\u91cf\u5904\u7406\u3001\u89d2\u8272\u8bbe\u8ba1\u7ba1\u7ebf）<\/li>\n
                                • \u672c\u6587\u914d\u5957\u6d4b\u8bd5\u7d20\u6750（10\u5f20\u53c2\u8003\u56fe+3\u4e2a\u63d0\u793a\u8bcd\u6a21\u677f）<\/li>\n<\/ul>\n

                                  8.3 \u5b98\u65b9\u8d44\u6e90<\/h4>\n
                                    \n
                                  • ComfyUI GitHub：https:\/\/github.com\/comfyanonymous\/ComfyUI\/releases\/tag\/v0.34.0<\/li>\n
                                  • \u5b98\u65b9\u6587\u6863：https:\/\/docs.comfyui.org\/<\/li>\n
                                  • \u793e\u533a\u5de5\u4f5c\u6d41\u5206\u4eab：https:\/\/comfyworkflows.com\/<\/li>\n<\/ul>\n
                                    \n

                                    \u4e5d\u3001\u603b\u7ed3<\/h3>\n

                                    ComfyUI v0.34.0\u7684\u66f4\u65b0\u903b\u8f91\u975e\u5e38\u6e05\u6670：\u4ece2D\u56fe\u50cf\u5de5\u5177\u8fdb\u5316\u4e3a3D+\u89c6\u9891+\u97f3\u9891\u7684\u5168\u5a92\u4f53\u751f\u4ea7\u5e73\u53f0\u3002\u5bf9\u521b\u4f5c\u8005\u6765\u8bf4，\u8fd9\u610f\u5473\u7740：<\/p>\n

                                  • 3D\u8bbe\u8ba1\u5e08：\u5355\u5f20\u6982\u5ff5\u56fe\u21923D\u8d44\u4ea7\u7684\u65f6\u95f4\u4ece\u6570\u5c0f\u65f6\u538b\u7f29\u5230\u5206\u949f\u7ea7<\/li>\n
                                  • \u89c6\u9891\u521b\u4f5c\u8005：HDR\u5bfc\u51fa\u8ba9AI\u751f\u6210\u89c6\u9891\u7684\u753b\u8d28\u9996\u6b21\u8fbe\u5230\u5546\u4e1a\u4ea4\u4ed8\u6807\u51c6<\/li>\n
                                  • \u97f3\u9891\u521b\u4f5c\u8005：MiniMax-H3\u7684\u97f3\u89c6\u9891\u4e00\u4f53\u5316，\u8ba9"\u914d\u97f3"\u4ece\u540e\u671f\u5de5\u4f5c\u53d8\u6210\u751f\u6210\u73af\u8282\u7684\u4e00\u90e8\u5206<\/li>\n
                                  • \u5f00\u53d1\u8005：OpenAI\u517c\u5bb9\u63a5\u53e3\u8ba9ComfyUI\u53ef\u4ee5\u65e0\u7f1d\u63a5\u5165\u73b0\u6709\u5de5\u5177\u94fe<\/li>\n

                                    \u5982\u679c\u4f60\u8fd8\u5728\u72b9\u8c6b\u662f\u5426\u5347\u7ea7，\u7b54\u6848\u662f\u80af\u5b9a\u7684\u3002 v0.34.0\u4e0d\u662f\u9526\u4e0a\u6dfb\u82b1，\u662fComfyUI\u4ece"AI\u7ed8\u753b\u5de5\u5177"\u5230"\u591a\u5a92\u4f53\u751f\u4ea7\u7ba1\u7ebf"\u7684\u5173\u952e\u4e00\u8dc3\u3002<\/p>\n

                                    \n

                                    \u672c\u6587\u6240\u6709\u5b9e\u6d4b\u6570\u636e\u57fa\u4e8eRTX 4090 24GB + Windows 11 + Python 3.11\u73af\u5883\u3002\u4e0d\u540c\u786c\u4ef6\u914d\u7f6e\u7684\u6027\u80fd\u8868\u73b0\u4f1a\u6709\u5dee\u5f02，\u6b22\u8fce\u5728\u8bc4\u8bba\u533a\u5206\u4eab\u4f60\u7684\u5b9e\u6d4b\u7ed3\u679c\u3002<\/p>\n

                                    \u4f5c\u8005：\u8d5b\u535a\u4ed3\u9f20
                                    \n\u5173\u6ce8\u6211\u53ef\u4ee5\u83b7\u53d6\u66f4\u591aAI\u5de5\u5177\u90e8\u7f72\u4e0e\u5b9e\u6218\u6307\u5357<\/p>\n","protected":false},"excerpt":{"rendered":"

                                    ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357：HDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41
                                    \n\u4f5c\u8005：\u8d5b\u535a\u4ed3\u9f20\u4e00\u53e5\u8bdd\u603b\u7ed3：ComfyUI v0.34.0\u4e0d\u662f\u4e00\u6b21\u5c0f\u4fee\u5c0f\u8865\u2014\u2014\u5b83\u628aHDR\u89c6\u9891\u5bfc\u51fa\u30013D\u8d44\u4ea7\u751f\u6210\u3001\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u8fd9\u4e09\u4e2a\u539f\u672c\u9700\u8981N\u4e2a\u8f6f\u4ef6\u624d\u80fd\u5b8c\u6210\u7684\u6d41\u7a0b，\u5168\u90e8\u585e\u8fdb\u4e86\u4e00\u4e2a\u8282\u70b9\u56fe\u91cc\u3002\u5982\u679c\u4f60\u8fd8\u5728\u7528v0.33.x，\u8fd9\u7bc7\u6307\u5357\u5c31\u662f\u5347\u7ea7\u7684\u7406\u7531\u3002\u4e00\u3001\u4e3a\u4ec0\u4e48\u4f60\u73b0\u5728\u5fc5\u987b\u5347\u7ea7\u5230v0.34.0<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[2477,66,52,1229,214],"topic":[],"class_list":["post-104935","post","type-post","status-publish","format-standard","hentry","category-server","tag-3d","tag-ai","tag-aigc","tag-214"],"yoast_head":"\nComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357\uff1aHDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.wsisp.com\/helps\/104935.html\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357\uff1aHDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"og:description\" content=\"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357：HDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41 \u4f5c\u8005：\u8d5b\u535a\u4ed3\u9f20\u4e00\u53e5\u8bdd\u603b\u7ed3：ComfyUI v0.34.0\u4e0d\u662f\u4e00\u6b21\u5c0f\u4fee\u5c0f\u8865\u2014\u2014\u5b83\u628aHDR\u89c6\u9891\u5bfc\u51fa\u30013D\u8d44\u4ea7\u751f\u6210\u3001\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u8fd9\u4e09\u4e2a\u539f\u672c\u9700\u8981N\u4e2a\u8f6f\u4ef6\u624d\u80fd\u5b8c\u6210\u7684\u6d41\u7a0b，\u5168\u90e8\u585e\u8fdb\u4e86\u4e00\u4e2a\u8282\u70b9\u56fe\u91cc\u3002\u5982\u679c\u4f60\u8fd8\u5728\u7528v0.33.x，\u8fd9\u7bc7\u6307\u5357\u5c31\u662f\u5347\u7ea7\u7684\u7406\u7531\u3002\u4e00\u3001\u4e3a\u4ec0\u4e48\u4f60\u73b0\u5728\u5fc5\u987b\u5347\u7ea7\u5230v0.34.0\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.wsisp.com\/helps\/104935.html\" \/>\n<meta property=\"og:site_name\" content=\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"article:published_time\" content=\"2026-09-13T05:25:23+00:00\" \/>\n<meta name=\"author\" content=\"admin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"7 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/104935.html\",\"url\":\"https:\/\/www.wsisp.com\/helps\/104935.html\",\"name\":\"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357\uff1aHDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"isPartOf\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\"},\"datePublished\":\"2026-09-13T05:25:23+00:00\",\"dateModified\":\"2026-09-13T05:25:23+00:00\",\"author\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/104935.html#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.wsisp.com\/helps\/104935.html\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/104935.html#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.wsisp.com\/helps\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357\uff1aHDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\",\"url\":\"https:\/\/www.wsisp.com\/helps\/\",\"name\":\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"description\":\"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\",\"name\":\"admin\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"contentUrl\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"caption\":\"admin\"},\"sameAs\":[\"http:\/\/wp.wsisp.com\"],\"url\":\"https:\/\/www.wsisp.com\/helps\/author\/admin\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357\uff1aHDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.wsisp.com\/helps\/104935.html","og_locale":"zh_CN","og_type":"article","og_title":"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357\uff1aHDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","og_description":"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357：HDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41 \u4f5c\u8005：\u8d5b\u535a\u4ed3\u9f20\u4e00\u53e5\u8bdd\u603b\u7ed3：ComfyUI v0.34.0\u4e0d\u662f\u4e00\u6b21\u5c0f\u4fee\u5c0f\u8865\u2014\u2014\u5b83\u628aHDR\u89c6\u9891\u5bfc\u51fa\u30013D\u8d44\u4ea7\u751f\u6210\u3001\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u8fd9\u4e09\u4e2a\u539f\u672c\u9700\u8981N\u4e2a\u8f6f\u4ef6\u624d\u80fd\u5b8c\u6210\u7684\u6d41\u7a0b，\u5168\u90e8\u585e\u8fdb\u4e86\u4e00\u4e2a\u8282\u70b9\u56fe\u91cc\u3002\u5982\u679c\u4f60\u8fd8\u5728\u7528v0.33.x，\u8fd9\u7bc7\u6307\u5357\u5c31\u662f\u5347\u7ea7\u7684\u7406\u7531\u3002\u4e00\u3001\u4e3a\u4ec0\u4e48\u4f60\u73b0\u5728\u5fc5\u987b\u5347\u7ea7\u5230v0.34.0","og_url":"https:\/\/www.wsisp.com\/helps\/104935.html","og_site_name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","article_published_time":"2026-09-13T05:25:23+00:00","author":"admin","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"admin","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"7 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.wsisp.com\/helps\/104935.html","url":"https:\/\/www.wsisp.com\/helps\/104935.html","name":"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357\uff1aHDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","isPartOf":{"@id":"https:\/\/www.wsisp.com\/helps\/#website"},"datePublished":"2026-09-13T05:25:23+00:00","dateModified":"2026-09-13T05:25:23+00:00","author":{"@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41"},"breadcrumb":{"@id":"https:\/\/www.wsisp.com\/helps\/104935.html#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.wsisp.com\/helps\/104935.html"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.wsisp.com\/helps\/104935.html#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.wsisp.com\/helps"},{"@type":"ListItem","position":2,"name":"ComfyUI v0.34.0 \u5b8c\u5168\u90e8\u7f72\u4e0e3D\u751f\u6210\u5b9e\u6218\u6307\u5357\uff1aHDR\u89c6\u9891\u3001TRELLIS2\u5efa\u6a21\u4e0eMiniMax-H3\u97f3\u89c6\u9891\u4e00\u4f53\u5316\u5de5\u4f5c\u6d41"}]},{"@type":"WebSite","@id":"https:\/\/www.wsisp.com\/helps\/#website","url":"https:\/\/www.wsisp.com\/helps\/","name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","description":"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41","name":"admin","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/","url":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","contentUrl":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","caption":"admin"},"sameAs":["http:\/\/wp.wsisp.com"],"url":"https:\/\/www.wsisp.com\/helps\/author\/admin"}]}},"_links":{"self":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/104935","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/comments?post=104935"}],"version-history":[{"count":0,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/104935\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/media?parent=104935"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/categories?post=104935"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/tags?post=104935"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/topic?post=104935"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}