{
  "1": {
    "class_type": "UNETLoader",
    "inputs": {
      "unet_name": "minimax_h3_ref2va_pruned_int8_convrot.safetensors",
      "weight_dtype": "default"
    }
  },
  "m0": {
    "class_type": "MiniMaxH3TurboLoRA",
    "inputs": {
      "lora_name": "minimax_h3_fl2v_lightx2v_turbo_8step_v1.0_comfy.safetensors",
      "strength": 1.0,
      "low_vram": false,
      "model": [
        "1",
        0
      ]
    },
    "_comment": "lightx2v Minimax-h3 Turbo **8step v1.0** 단독 @1.0 (2026-08-12 `t_4e40bb6cf50d` · CEO 승격 지시). 구 `ckpt850`(4step) 에서 교체 — 격자 4판 실측에서 **8step·캐시ON 이 최적**이었다: 6step·캐시OFF 대비 **80초 빠르고**(521~525s vs 601s) 오디오·화질이 같거나 낫다. ★6step·캐시ON 은 **오디오가 갈라진다**(CEO 육안 \"목소리 갈라지네\") — 속도만 보고 6step 으로 내리지 말 것."
  },
  "L2": {
    "class_type": "LoraLoaderModelOnly",
    "inputs": {
      "lora_name": "MysticXXX_MMH3-V4.safetensors",
      "strength_model": 0.4,
      "model": [
        "m0",
        0
      ]
    },
    "_comment": "Mystic XXX `MysticXXX_MMH3-V4` **0.4** — 2026-09-06 `t_0ae98911f9cb` 신설 슬롯. 체인은 `m0(터보) → L2(mystic 0.4) → m1(sage)`. ★★2026-09-10 CEO 판정으로 **realism(구 L1) 을 이 판에서 걷었다**(`t_2ac5b03c5db2`) — 사유 = *\"mystic 과 realism 이 같이 적용되면 영상 품질이 망가지더라. 각각 쓰면 괜찮은데\"*. ⇒ 종전 이 자리에 있던 *\"realism 을 빼고 그 칸을 쓰지 않았다\"* 는 **그 판정 이전의 입장**이라 걷었다 (이 필드는 실행 `workflow.json` 에까지 실리므로, 낡은 입장을 현재형으로 남기면 읽는 사람이 **누락으로 읽고 되넣는다**). ★realism 자체가 기각된 게 아니라 **Mystic 과의 동시 적용**이 기각됐다. ★값 0.4 의 근거 = 형제 `ref2va-minimax-h3-mystic-5090-1` 의 저작값과 같게 둔다(같은 파일을 두 판이 다른 강도로 쓰면 「어느 쪽이 정본인가」가 생긴다). 제작자 권장은 0.5~0.9 라 우리 값은 그 아래이고, 강도 확정은 CEO 육안 별건이다. ★끄려면 `spec_override.toggles {\"lora_2\": false}` — 그러면 러너 splice 가 `m1.model` 을 `[\"m0\", 0]` 로 되돌려 **LoRA 0장(터보만)** 체인이 된다."
  },
  "m1": {
    "class_type": "MiniMaxH3MemoryEfficientSageAttentionPatch",
    "inputs": {
      "model": [
        "L2",
        0
      ]
    }
  },
  "2": {
    "class_type": "CLIPLoader",
    "inputs": {
      "clip_name": "qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors",
      "type": "minimax",
      "device": "default"
    }
  },
  "3": {
    "class_type": "VAELoader",
    "inputs": {
      "vae_name": "minimax_h3_video_vae_fp16.safetensors"
    }
  },
  "4": {
    "class_type": "VAELoader",
    "inputs": {
      "vae_name": "minimax_h3_audio_vae_fp32.safetensors"
    }
  },
  "101": {
    "class_type": "LoadImage",
    "inputs": {
      "image": "dockaiq-input-460510-input_image-373b128b-4135-48ab-ba6b-d2284eb23fe3.webp"
    }
  },
  "102": {
    "class_type": "LoadImage",
    "inputs": {
      "image": "dockaiq-input-460510-picture_2-ref-3shot.webp"
    }
  },
  "103": {
    "class_type": "LoadImage",
    "inputs": {
      "image": "dockaiq-input-460510-picture_3-3880888d-5900-4a22-9749-3f92d848a988-outer.webp"
    }
  },
  "104": {
    "class_type": "LoadImage",
    "inputs": {
      "image": "dockaiq-input-460510-picture_4-fe29be2e-7395-4eb8-90bc-80feaa9b8bd8-under.webp"
    }
  },
  "6": {
    "class_type": "MiniMaxH3ReferenceToVideo",
    "inputs": {
      "prompt": "<Picture 1> is the location (this room); keep its layout and furnishings, surfaces clear of small objects.\n<Picture 2> is <Subject 1> — this exact person; keep the same face and hair. <Picture 3> is <Subject 1>'s clothing sheet. <Picture 4> is <Subject 1>'s underwear sheet.\n\n<Subject 1> is the woman with dark brown long wavy hair in <Picture 2>. She is standing at the foot of the bed, center of frame, close to the viewer, turning to face the lens with her hands lowering from the ribbon at her back, wearing the black sheer chiffon maid dress zipped up at the back, the white frilly half apron tied at the waist with the ribbon knotted at the back, the white lace maid headband set on her head, the black sheer thigh-high stockings pulled up from the toes over the calves, and the black patent mary jane heels in <Picture 3>.\n\nThe viewer, a man, is standing close behind her at the foot of the bed, seeing this room through his own eyes. His face, head, torso and hips are never in frame — the camera is where he is. He does not touch her in this shot.\n\n[Shot 1] First-person POV from the viewer's own eyes, at his standing eye height, close to <Subject 1> and facing her as she turns toward the lens. She stands at the foot of the bed, center of frame, then turns to face the camera and takes half a step toward the window; the camera follows her only as much as the viewer's head actually turns and tilts, staying at his eye position. She speaks to the viewer, looking into the camera, teasing and a little playful, with a slight smile audible in her voice: <d>[Korean] 이제 진짜 처음이네요.</d> As she speaks she shakes out the hem of her apron with both hands. Then she takes half a step toward the window, and speaking to the viewer, looking into the camera, warm and matter-of-fact: <d>[Korean] 마지막은 서서 한 손은 앞치마 주머니에 넣고 창밖을 볼게요.</d> As she speaks she slides one hand into the apron pocket and looks out the window, then lifts her chin slightly and turns her eyes toward the daylight coming through the window. <Subject 1> (<Picture 2>) is the ONLY one who speaks in this shot, speaking to the viewer.\n\nMatch visible mouth motion to each spoken phrase, then let the mouth rest naturally between phrases. Speak these two lines in order as one flowing thought. Use natural breath spacing, one brief hesitation, and a tiny laugh only if it fits. Do not repeat, rewrite, add, sing, chant, or narrate the words.\n\nSubtle early reflections from nearby walls. The dialogue is naturally embedded in the environment rather than sounding close-miked or studio-recorded. Soft environmental ambience continuously fills the space. Very subtle natural reverberation. No artificial narration quality.\n\nNo text overlays, no subtitles, no watermark, no extra people in the background, no camera shake, no jump cuts, no music, no lyrics, no singing, no narration.",
      "width": 512,
      "height": 896,
      "length": 243,
      "ref_image_size": "match",
      "clip": [
        "2",
        0
      ],
      "vae": [
        "3",
        0
      ],
      "audio_vae": [
        "4",
        0
      ],
      "ref_images.ref_image_0": [
        "101",
        0
      ],
      "ref_images.ref_image_1": [
        "102",
        0
      ],
      "ref_images.ref_image_2": [
        "103",
        0
      ],
      "ref_images.ref_image_3": [
        "104",
        0
      ]
    }
  },
  "7": {
    "class_type": "BasicGuider",
    "inputs": {
      "model": [
        "mS",
        0
      ],
      "conditioning": [
        "6",
        0
      ]
    }
  },
  "8": {
    "inputs": {
      "sampler_name": "er_sde"
    },
    "class_type": "KSamplerSelect",
    "_meta": {
      "title": "KSampler (선택)"
    }
  },
  "9": {
    "class_type": "BasicScheduler",
    "inputs": {
      "scheduler": "beta",
      "steps": 8,
      "denoise": 1.0,
      "model": [
        "mS",
        0
      ]
    }
  },
  "10": {
    "class_type": "RandomNoise",
    "inputs": {
      "noise_seed": 1398894998
    }
  },
  "11": {
    "class_type": "SamplerCustomAdvanced",
    "inputs": {
      "noise": [
        "10",
        0
      ],
      "guider": [
        "7",
        0
      ],
      "sampler": [
        "8",
        0
      ],
      "sigmas": [
        "9",
        0
      ],
      "latent_image": [
        "6",
        1
      ]
    }
  },
  "12": {
    "class_type": "VAEDecode",
    "inputs": {
      "samples": [
        "11",
        0
      ],
      "vae": [
        "vb",
        0
      ]
    }
  },
  "13": {
    "class_type": "VAEDecodeAudio",
    "inputs": {
      "samples": [
        "11",
        0
      ],
      "vae": [
        "4",
        0
      ]
    }
  },
  "14": {
    "class_type": "CreateVideo",
    "inputs": {
      "fps": 24,
      "bit_depth": 8,
      "images": [
        "wm",
        0
      ],
      "audio": [
        "13",
        0
      ]
    }
  },
  "15": {
    "class_type": "SaveVideo",
    "inputs": {
      "filename_prefix": "h3_r2v_v384_j460510",
      "format": "auto",
      "video": [
        "14",
        0
      ],
      "codec": "auto"
    }
  },
  "vb": {
    "class_type": "H3VideoVAEBatchDecode",
    "inputs": {
      "vae": [
        "3",
        0
      ],
      "enabled": true
    }
  },
  "mL": {
    "inputs": {
      "head_chunks": 4,
      "model": [
        "m1",
        0
      ]
    },
    "class_type": "MiniMaxLowVRAMAttention",
    "_meta": {
      "title": "MiniMax H3 Low VRAM Attention"
    }
  },
  "mC": {
    "inputs": {
      "chunks": 2,
      "seq_threshold": 4096,
      "model": [
        "mL",
        0
      ]
    },
    "class_type": "MiniMaxChunkFeedForward",
    "_meta": {
      "title": "MiniMax H3 Chunk FeedForward"
    }
  },
  "mS": {
    "inputs": {
      "shift_video": 12.0,
      "shift_audio": 6.0,
      "model": [
        "mC",
        0
      ]
    },
    "class_type": "MiniMaxH3SigmaShift",
    "_meta": {
      "title": "ModelSamplingMiniMaxH3"
    }
  },
  "wm": {
    "inputs": {
      "images": [
        "12",
        0
      ],
      "line1": "AI",
      "line2": "yusti.net",
      "height_px": 12,
      "alpha": 0.85,
      "margin_px": 13,
      "shadow_px": 1,
      "line3": "3e32f51e"
    },
    "class_type": "DockContentWatermark",
    "_meta": {
      "title": "법정 표시 워터마크 (t_88479c752c44)"
    }
  }
}