{
  "id": "juggz-img2img-customer-0001",
  "revision": 0,
  "last_node_id": 1111,
  "last_link_id": 12,
  "nodes": [
    {
      "id": 1001,
      "type": "Note",
      "pos": [
        80,
        330
      ],
      "size": [
        430,
        180
      ],
      "flags": {},
      "order": 0,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ① Load Model — Juggernaut Z",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "MODEL — Juggernaut Z\n\nLoads the Juggernaut Z model (a fine-tune of the Z-Image base). On RunDiffusion it lives in the diffusion_models folder, so it loads as a UNET/diffusion model, NOT as a checkpoint.\n\nSelect: juggernautZ_v10ByRundiffusion.safetensors  (the public v1.0 release)."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 1,
      "type": "UNETLoader",
      "pos": [
        80,
        535
      ],
      "size": [
        430,
        110
      ],
      "flags": {},
      "order": 1,
      "mode": 0,
      "inputs": [],
      "outputs": [
        {
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            3
          ],
          "slot_index": 0
        }
      ],
      "title": "① Load Model — Juggernaut Z",
      "properties": {
        "Node name for S&R": "UNETLoader"
      },
      "widgets_values": [
        "juggernautZ_v10ByRundiffusion.safetensors",
        "default"
      ]
    },
    {
      "id": 1002,
      "type": "Note",
      "pos": [
        80,
        765
      ],
      "size": [
        430,
        180
      ],
      "flags": {},
      "order": 2,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ② Load Text Encoder — Qwen3-4B",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "TEXT ENCODER — Qwen3-4B\n\nTurns your text prompt into guidance the model understands.\nIMPORTANT: the 'type' MUST be set to 'qwen_image' for Z-Image / Juggernaut Z. Any other type will fail to load this encoder."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 2,
      "type": "CLIPLoader",
      "pos": [
        80,
        970
      ],
      "size": [
        430,
        130
      ],
      "flags": {},
      "order": 3,
      "mode": 0,
      "inputs": [],
      "outputs": [
        {
          "name": "CLIP",
          "type": "CLIP",
          "links": [
            1,
            2
          ],
          "slot_index": 0
        }
      ],
      "title": "② Load Text Encoder — Qwen3-4B",
      "properties": {
        "Node name for S&R": "CLIPLoader"
      },
      "widgets_values": [
        "qwen_3_4b.safetensors",
        "qwen_image",
        "default"
      ]
    },
    {
      "id": 1003,
      "type": "Note",
      "pos": [
        80,
        1220
      ],
      "size": [
        430,
        180
      ],
      "flags": {},
      "order": 4,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ③ Load VAE — ae.safetensors",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "VAE — ae.safetensors\n\nConverts between pixels and the model's latent space. Used twice: to ENCODE your source image (step 5) and to DECODE the result (step 10)."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 3,
      "type": "VAELoader",
      "pos": [
        80,
        1425
      ],
      "size": [
        430,
        90
      ],
      "flags": {},
      "order": 5,
      "mode": 0,
      "inputs": [],
      "outputs": [
        {
          "name": "VAE",
          "type": "VAE",
          "links": [
            8,
            11
          ],
          "slot_index": 0
        }
      ],
      "title": "③ Load VAE — ae.safetensors",
      "properties": {
        "Node name for S&R": "VAELoader"
      },
      "widgets_values": [
        "ae.safetensors"
      ]
    },
    {
      "id": 1004,
      "type": "Note",
      "pos": [
        80,
        1635
      ],
      "size": [
        430,
        180
      ],
      "flags": {},
      "order": 6,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ④ Source Image (your img2img input)",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "SOURCE IMAGE (img2img input)\n\nThis is the image you want to transform. Click to upload or pick your own. Everything downstream reshapes THIS image according to your prompt + denoise."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 4,
      "type": "LoadImage",
      "pos": [
        80,
        1840
      ],
      "size": [
        430,
        400
      ],
      "flags": {},
      "order": 7,
      "mode": 0,
      "inputs": [],
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            7
          ],
          "slot_index": 0
        },
        {
          "name": "MASK",
          "type": "MASK",
          "links": [],
          "slot_index": 1
        }
      ],
      "title": "④ Source Image (your img2img input)",
      "properties": {
        "Node name for S&R": "LoadImage"
      },
      "widgets_values": [
        "Juggernaut_Z_00062_.png",
        "image"
      ]
    },
    {
      "id": 1008,
      "type": "Note",
      "pos": [
        700,
        330
      ],
      "size": [
        380,
        180
      ],
      "flags": {},
      "order": 8,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ⑧ Model Sampling (shift)",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "MODEL SAMPLING (shift)\n\nSets the flow-model sampling 'shift' that Z-Image needs. ~3.0 is a solid default. Higher shift = a bit more contrast/structure; lower = softer."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 8,
      "type": "ModelSamplingAuraFlow",
      "pos": [
        700,
        535
      ],
      "size": [
        380,
        100
      ],
      "flags": {},
      "order": 9,
      "mode": 0,
      "inputs": [
        {
          "name": "model",
          "type": "MODEL",
          "link": 3
        }
      ],
      "outputs": [
        {
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            4
          ],
          "slot_index": 0
        }
      ],
      "title": "⑧ Model Sampling (shift)",
      "properties": {
        "Node name for S&R": "ModelSamplingAuraFlow"
      },
      "widgets_values": [
        3.0
      ]
    },
    {
      "id": 1006,
      "type": "Note",
      "pos": [
        700,
        755
      ],
      "size": [
        460,
        180
      ],
      "flags": {},
      "order": 10,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ⑥ Positive Prompt (what you WANT)",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "POSITIVE PROMPT\n\nDescribe what you WANT to see. Juggernaut Z rewards clear, visual, intentional prompts — choose the right words rather than piling on more."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 6,
      "type": "TextEncodeZImageOmni",
      "pos": [
        700,
        960
      ],
      "size": [
        460,
        260
      ],
      "flags": {},
      "order": 11,
      "mode": 0,
      "inputs": [
        {
          "name": "clip",
          "type": "CLIP",
          "link": 1
        }
      ],
      "outputs": [
        {
          "name": "CONDITIONING",
          "type": "CONDITIONING",
          "links": [
            5
          ],
          "slot_index": 0
        }
      ],
      "title": "⑥ Positive Prompt (what you WANT)",
      "properties": {
        "Node name for S&R": "TextEncodeZImageOmni"
      },
      "widgets_values": [
        "a young woman with long dark blue hair and piercing blue eyes, wearing a black hoodie, cinematic lighting, ultra detailed skin, sharp focus, photorealistic",
        true
      ]
    },
    {
      "id": 1007,
      "type": "Note",
      "pos": [
        700,
        1340
      ],
      "size": [
        460,
        180
      ],
      "flags": {},
      "order": 12,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ⑦ Negative Prompt (what you DON'T want)",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "NEGATIVE PROMPT\n\nDescribe what you DON'T want (blur, artifacts, extra limbs, etc.). With CFG 6 this meaningfully steers the result away from these."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 7,
      "type": "TextEncodeZImageOmni",
      "pos": [
        700,
        1545
      ],
      "size": [
        460,
        220
      ],
      "flags": {},
      "order": 13,
      "mode": 0,
      "inputs": [
        {
          "name": "clip",
          "type": "CLIP",
          "link": 2
        }
      ],
      "outputs": [
        {
          "name": "CONDITIONING",
          "type": "CONDITIONING",
          "links": [
            6
          ],
          "slot_index": 0
        }
      ],
      "title": "⑦ Negative Prompt (what you DON'T want)",
      "properties": {
        "Node name for S&R": "TextEncodeZImageOmni"
      },
      "widgets_values": [
        "blurry, distorted, ugly, low resolution, malformed, grainy, error, jpeg artifacts, duplicate, poorly drawn",
        true
      ]
    },
    {
      "id": 1005,
      "type": "Note",
      "pos": [
        700,
        1885
      ],
      "size": [
        300,
        180
      ],
      "flags": {},
      "order": 14,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ⑤ Encode Image → Latent",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "ENCODE IMAGE → LATENT\n\nConverts your source image into a latent the sampler can work on. This step is what makes the graph img2img instead of text-to-image."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 5,
      "type": "VAEEncode",
      "pos": [
        700,
        2090
      ],
      "size": [
        300,
        100
      ],
      "flags": {},
      "order": 15,
      "mode": 0,
      "inputs": [
        {
          "name": "pixels",
          "type": "IMAGE",
          "link": 7
        },
        {
          "name": "vae",
          "type": "VAE",
          "link": 8
        }
      ],
      "outputs": [
        {
          "name": "LATENT",
          "type": "LATENT",
          "links": [
            9
          ],
          "slot_index": 0
        }
      ],
      "title": "⑤ Encode Image → Latent",
      "properties": {
        "Node name for S&R": "VAEEncode"
      },
      "widgets_values": []
    },
    {
      "id": 1009,
      "type": "Note",
      "pos": [
        1320,
        330
      ],
      "size": [
        380,
        180
      ],
      "flags": {},
      "order": 16,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ⑨ Sampler — Juggernaut Z settings",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "SAMPLER — the engine\n\nJuggernaut Z recommended: CFG 6 (range 6-9), 35 steps (25-45), sampler res_multistep, scheduler beta.\n\nDENOISE = img2img strength:\n  0.4  subtle touch-up\n  0.65 balanced (default)\n  0.8  strong restyle\nLower keeps your source; higher follows the prompt."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 9,
      "type": "KSampler",
      "pos": [
        1320,
        535
      ],
      "size": [
        380,
        300
      ],
      "flags": {},
      "order": 17,
      "mode": 0,
      "inputs": [
        {
          "name": "model",
          "type": "MODEL",
          "link": 4
        },
        {
          "name": "positive",
          "type": "CONDITIONING",
          "link": 5
        },
        {
          "name": "negative",
          "type": "CONDITIONING",
          "link": 6
        },
        {
          "name": "latent_image",
          "type": "LATENT",
          "link": 9
        }
      ],
      "outputs": [
        {
          "name": "LATENT",
          "type": "LATENT",
          "links": [
            10
          ],
          "slot_index": 0
        }
      ],
      "title": "⑨ Sampler — Juggernaut Z settings",
      "properties": {
        "Node name for S&R": "KSampler"
      },
      "widgets_values": [
        808520384108623,
        "randomize",
        35,
        6.0,
        "res_multistep",
        "beta",
        0.65
      ]
    },
    {
      "id": 1010,
      "type": "Note",
      "pos": [
        1820,
        330
      ],
      "size": [
        300,
        180
      ],
      "flags": {},
      "order": 18,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ⑩ Decode Latent → Image",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "DECODE LATENT → IMAGE\n\nTurns the finished latent back into a normal viewable image using the VAE."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 10,
      "type": "VAEDecode",
      "pos": [
        1820,
        535
      ],
      "size": [
        300,
        100
      ],
      "flags": {},
      "order": 19,
      "mode": 0,
      "inputs": [
        {
          "name": "samples",
          "type": "LATENT",
          "link": 10
        },
        {
          "name": "vae",
          "type": "VAE",
          "link": 11
        }
      ],
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            12
          ],
          "slot_index": 0
        }
      ],
      "title": "⑩ Decode Latent → Image",
      "properties": {
        "Node name for S&R": "VAEDecode"
      },
      "widgets_values": []
    },
    {
      "id": 1011,
      "type": "Note",
      "pos": [
        2180,
        330
      ],
      "size": [
        460,
        180
      ],
      "flags": {},
      "order": 20,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "ℹ ⑪ Save Image",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "SAVE IMAGE\n\nWrites the final image to your outputs. Change the filename prefix (e.g. MyProject/%date) to organize your renders."
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 11,
      "type": "SaveImage",
      "pos": [
        2180,
        535
      ],
      "size": [
        460,
        460
      ],
      "flags": {},
      "order": 21,
      "mode": 0,
      "inputs": [
        {
          "name": "images",
          "type": "IMAGE",
          "link": 12
        }
      ],
      "outputs": [],
      "title": "⑪ Save Image",
      "properties": {
        "Node name for S&R": "SaveImage"
      },
      "widgets_values": [
        "JuggernautZ-Img2Img"
      ]
    },
    {
      "id": 100,
      "type": "Note",
      "pos": [
        80,
        60
      ],
      "size": [
        1100,
        210
      ],
      "flags": {},
      "order": 22,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "README — read me first",
      "properties": {
        "text": ""
      },
      "widgets_values": [
        "JUGGERNAUT Z — IMG2IMG  (RunDiffusion / ComfyUI)\n\nTransform any source image with the Juggernaut Z model.\nQUICK START:\n  1) Set your image in node ④.\n  2) Edit the Positive ⑥ and Negative ⑦ prompts.\n  3) Set DENOISE in node ⑨ (0.65 = balanced).\n  4) Queue Prompt.\n\nFollow the numbered steps left → right. The ℹ notes explain each node."
      ],
      "color": "#225",
      "bgcolor": "#447"
    }
  ],
  "links": [
    [
      1,
      2,
      0,
      6,
      0,
      "CLIP"
    ],
    [
      2,
      2,
      0,
      7,
      0,
      "CLIP"
    ],
    [
      3,
      1,
      0,
      8,
      0,
      "MODEL"
    ],
    [
      4,
      8,
      0,
      9,
      0,
      "MODEL"
    ],
    [
      5,
      6,
      0,
      9,
      1,
      "CONDITIONING"
    ],
    [
      6,
      7,
      0,
      9,
      2,
      "CONDITIONING"
    ],
    [
      7,
      4,
      0,
      5,
      0,
      "IMAGE"
    ],
    [
      8,
      3,
      0,
      5,
      1,
      "VAE"
    ],
    [
      9,
      5,
      0,
      9,
      3,
      "LATENT"
    ],
    [
      10,
      9,
      0,
      10,
      0,
      "LATENT"
    ],
    [
      11,
      3,
      0,
      10,
      1,
      "VAE"
    ],
    [
      12,
      10,
      0,
      11,
      0,
      "IMAGE"
    ]
  ],
  "groups": [
    {
      "id": 1,
      "title": "Juggernaut Z — IMG to IMG (annotated for customers)",
      "bounding": [
        40,
        30,
        2660,
        2370
      ],
      "color": "#3f789e",
      "font_size": 28,
      "flags": {}
    }
  ],
  "config": {},
  "extra": {
    "ds": {
      "scale": 0.55,
      "offset": [
        0,
        0
      ]
    }
  },
  "version": 0.4
}