{
  "last_node_id": 14,
  "last_link_id": 16,
  "nodes": [
    {
      "id": 1,
      "type": "UNETLoader",
      "pos": [
        40,
        60
      ],
      "size": [
        315,
        82
      ],
      "flags": {},
      "order": 0,
      "mode": 0,
      "inputs": [],
      "outputs": [
        {
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            3,
            4
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "UNETLoader"
      },
      "widgets_values": [
        "moss_image_sft_int8_convrot.safetensors",
        "default"
      ]
    },
    {
      "id": 2,
      "type": "CLIPLoader",
      "pos": [
        40,
        200
      ],
      "size": [
        315,
        106
      ],
      "flags": {},
      "order": 1,
      "mode": 0,
      "inputs": [],
      "outputs": [
        {
          "name": "CLIP",
          "type": "CLIP",
          "links": [
            1,
            2
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "CLIPLoader"
      },
      "widgets_values": [
        "qwen3.5_9b_w4a8_gptq.safetensors",
        "moss_image",
        "default"
      ]
    },
    {
      "id": 3,
      "type": "CLIPTextEncode",
      "pos": [
        400,
        340
      ],
      "size": [
        440,
        240
      ],
      "flags": {},
      "order": 3,
      "mode": 0,
      "inputs": [
        {
          "name": "clip",
          "type": "CLIP",
          "link": 1
        }
      ],
      "outputs": [
        {
          "name": "CONDITIONING",
          "type": "CONDITIONING",
          "links": [
            7,
            11
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "CLIPTextEncode"
      },
      "widgets_values": [
        "Anime-style digital illustration of an adult woman in her late twenties closing her small flower shop on a rainy night. She is carrying the last zinc bucket of red tulips in from the pavement, both hands around its wire handle, and nudging the half-open glass door wider with her right shoulder; her eyes are on the doorstep at her feet. She has a short dark bob tucked behind one ear and wears a charcoal canvas apron over a cream knit sweater with the sleeves pushed up.\n\nThe view is a medium shot that frames her from the knees up, from the wet pavement at waist height a little to her left, so the lit shop fills the frame behind her: shelves of potted plants, galvanized buckets and a long wooden worktable. The warm light of the shop's ceiling lamps falls through the glass door onto her face, her hands and the tulips, and spills in a long rectangle across the wet pavement; outside that light the street is blue-grey and the rain shows as fine slanting lines against the dark buildings.\n\nDrawn in the look of a current anime television series: crisp, thin line art; a modern anime face with an adult's proportions, large detailed eyes with gradient irises and layered highlights, a small nose and a narrow chin; glossy hair with fine strand lines and a soft gradient; two-tone cel shading on skin and clothes; and a painted background with crisp perspective. The palette is muted blue-grey outside and warm, low-saturation ochre inside, so the red tulips are the only saturated color. The mood is quiet and tired, the end of a long day. No visible text."
      ],
      "title": "Positive"
    },
    {
      "id": 4,
      "type": "CLIPTextEncode",
      "pos": [
        400,
        620
      ],
      "size": [
        440,
        140
      ],
      "flags": {},
      "order": 4,
      "mode": 0,
      "inputs": [
        {
          "name": "clip",
          "type": "CLIP",
          "link": 2
        }
      ],
      "outputs": [
        {
          "name": "CONDITIONING",
          "type": "CONDITIONING",
          "links": [
            8,
            12
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "CLIPTextEncode"
      },
      "widgets_values": [
        "worst quality, bad anatomy, watermark, jpeg artifact, photorealistic, real person, selfie, photograph, 3d render, signature, artist name, extra fingers, fused fingers, missing fingers, extra hands, extra arms, malformed hands"
      ],
      "title": "Negative"
    },
    {
      "id": 5,
      "type": "EmptyFlux2LatentImage",
      "pos": [
        400,
        800
      ],
      "size": [
        315,
        106
      ],
      "flags": {},
      "order": 5,
      "mode": 0,
      "inputs": [],
      "outputs": [
        {
          "name": "LATENT",
          "type": "LATENT",
          "links": [
            9
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "EmptyFlux2LatentImage"
      },
      "widgets_values": [
        832,
        1248,
        1
      ]
    },
    {
      "id": 6,
      "type": "ModelSamplingAuraFlow",
      "pos": [
        760,
        60
      ],
      "size": [
        315,
        58
      ],
      "flags": {},
      "order": 6,
      "mode": 0,
      "inputs": [
        {
          "name": "model",
          "type": "MODEL",
          "link": 3
        }
      ],
      "outputs": [
        {
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            6
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "ModelSamplingAuraFlow"
      },
      "widgets_values": [
        5.0
      ],
      "title": "Shift for stage 1"
    },
    {
      "id": 7,
      "type": "KSamplerAdvanced",
      "pos": [
        1140,
        60
      ],
      "size": [
        315,
        334
      ],
      "flags": {},
      "order": 9,
      "mode": 0,
      "inputs": [
        {
          "name": "model",
          "type": "MODEL",
          "link": 6
        },
        {
          "name": "positive",
          "type": "CONDITIONING",
          "link": 7
        },
        {
          "name": "negative",
          "type": "CONDITIONING",
          "link": 8
        },
        {
          "name": "latent_image",
          "type": "LATENT",
          "link": 9
        }
      ],
      "outputs": [
        {
          "name": "LATENT",
          "type": "LATENT",
          "links": [
            13
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "KSamplerAdvanced"
      },
      "widgets_values": [
        "enable",
        2026102971,
        "fixed",
        12,
        2.5,
        "er_sde",
        "normal",
        0,
        4,
        "enable"
      ],
      "title": "Stage 1: base, steps 0 to 4"
    },
    {
      "id": 8,
      "type": "LoraLoaderModelOnly",
      "pos": [
        400,
        200
      ],
      "size": [
        315,
        82
      ],
      "flags": {},
      "order": 7,
      "mode": 0,
      "inputs": [
        {
          "name": "model",
          "type": "MODEL",
          "link": 4
        }
      ],
      "outputs": [
        {
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            5
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "LoraLoaderModelOnly"
      },
      "widgets_values": [
        "moss_turbolora_comfy.safetensors",
        1.0
      ],
      "title": "TurboLoRA, key-fixed file"
    },
    {
      "id": 9,
      "type": "ModelSamplingAuraFlow",
      "pos": [
        760,
        200
      ],
      "size": [
        315,
        58
      ],
      "flags": {},
      "order": 8,
      "mode": 0,
      "inputs": [
        {
          "name": "model",
          "type": "MODEL",
          "link": 5
        }
      ],
      "outputs": [
        {
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            10
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "ModelSamplingAuraFlow"
      },
      "widgets_values": [
        5.0
      ],
      "title": "Shift for stage 2"
    },
    {
      "id": 10,
      "type": "KSamplerAdvanced",
      "pos": [
        1500,
        60
      ],
      "size": [
        315,
        334
      ],
      "flags": {},
      "order": 10,
      "mode": 0,
      "inputs": [
        {
          "name": "model",
          "type": "MODEL",
          "link": 10
        },
        {
          "name": "positive",
          "type": "CONDITIONING",
          "link": 11
        },
        {
          "name": "negative",
          "type": "CONDITIONING",
          "link": 12
        },
        {
          "name": "latent_image",
          "type": "LATENT",
          "link": 13
        }
      ],
      "outputs": [
        {
          "name": "LATENT",
          "type": "LATENT",
          "links": [
            14
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "KSamplerAdvanced"
      },
      "widgets_values": [
        "disable",
        2026102971,
        "fixed",
        12,
        1.0,
        "euler",
        "normal",
        4,
        12,
        "disable"
      ],
      "title": "Stage 2: TurboLoRA, steps 4 to 12"
    },
    {
      "id": 11,
      "type": "VAELoader",
      "pos": [
        40,
        360
      ],
      "size": [
        315,
        58
      ],
      "flags": {},
      "order": 2,
      "mode": 0,
      "inputs": [],
      "outputs": [
        {
          "name": "VAE",
          "type": "VAE",
          "links": [
            15
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "VAELoader"
      },
      "widgets_values": [
        "flux2-vae_small.safetensors"
      ]
    },
    {
      "id": 12,
      "type": "VAEDecode",
      "pos": [
        1860,
        60
      ],
      "size": [
        210,
        46
      ],
      "flags": {},
      "order": 11,
      "mode": 0,
      "inputs": [
        {
          "name": "samples",
          "type": "LATENT",
          "link": 14
        },
        {
          "name": "vae",
          "type": "VAE",
          "link": 15
        }
      ],
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            16
          ],
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "VAEDecode"
      },
      "widgets_values": []
    },
    {
      "id": 13,
      "type": "SaveImage",
      "pos": [
        1860,
        160
      ],
      "size": [
        420,
        480
      ],
      "flags": {},
      "order": 12,
      "mode": 0,
      "inputs": [
        {
          "name": "images",
          "type": "IMAGE",
          "link": 16
        }
      ],
      "outputs": [],
      "properties": {
        "Node name for S&R": "SaveImage"
      },
      "widgets_values": [
        "moss_hybrid"
      ]
    },
    {
      "id": 14,
      "type": "MarkdownNote",
      "pos": [
        40,
        480
      ],
      "size": [
        320,
        460
      ],
      "flags": {},
      "order": 13,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "Read this first",
      "properties": {},
      "widgets_values": [
        "# MOSS Image hybrid 4 + 8\n\nStage 1 runs the first 4 steps on the base model. Stage 2 runs the last 8 steps with the TurboLoRA. Keep noise_seed the same in both.\n\n**The TurboLoRA does not work as published.** ComfyUI 0.42.1 maps 0 of its 524 keys. Run the short script in the article, then put moss_turbolora_comfy.safetensors in your loras folder.\n\nArticle with the script and the test results\nhttps://civitai.com/articles/38059\n\nFiles from Comfy-Org/MOSS-Image on Hugging Face\n- diffusion_models/moss_image_sft_int8_convrot.safetensors\n- text_encoders/qwen3.5_9b_w4a8_gptq.safetensors\n- vae/flux2-vae_small.safetensors\n\nTurboLoRA on ModelScope\nhttps://www.modelscope.cn/models/DiffSynth-Studio/MOSS-Image-SFT-TurboLoRA"
      ]
    }
  ],
  "links": [
    [
      1,
      2,
      0,
      3,
      0,
      "CLIP"
    ],
    [
      2,
      2,
      0,
      4,
      0,
      "CLIP"
    ],
    [
      3,
      1,
      0,
      6,
      0,
      "MODEL"
    ],
    [
      4,
      1,
      0,
      8,
      0,
      "MODEL"
    ],
    [
      5,
      8,
      0,
      9,
      0,
      "MODEL"
    ],
    [
      6,
      6,
      0,
      7,
      0,
      "MODEL"
    ],
    [
      7,
      3,
      0,
      7,
      1,
      "CONDITIONING"
    ],
    [
      8,
      4,
      0,
      7,
      2,
      "CONDITIONING"
    ],
    [
      9,
      5,
      0,
      7,
      3,
      "LATENT"
    ],
    [
      10,
      9,
      0,
      10,
      0,
      "MODEL"
    ],
    [
      11,
      3,
      0,
      10,
      1,
      "CONDITIONING"
    ],
    [
      12,
      4,
      0,
      10,
      2,
      "CONDITIONING"
    ],
    [
      13,
      7,
      0,
      10,
      3,
      "LATENT"
    ],
    [
      14,
      10,
      0,
      12,
      0,
      "LATENT"
    ],
    [
      15,
      11,
      0,
      12,
      1,
      "VAE"
    ],
    [
      16,
      12,
      0,
      13,
      0,
      "IMAGE"
    ]
  ],
  "groups": [],
  "config": {},
  "extra": {},
  "version": 0.4
}
