Duplicate from LiconStudio/LTX-2.3-Multiple-Subject-Reference

Co-authored-by: LiconStudio <LiconStudio@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,74 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+2_1.png filter=lfs diff=lfs merge=lfs -text
+2_2.png filter=lfs diff=lfs merge=lfs -text
+2_3.png filter=lfs diff=lfs merge=lfs -text
+2_4.png filter=lfs diff=lfs merge=lfs -text
+2_video.mp4 filter=lfs diff=lfs merge=lfs -text
+1_1.png filter=lfs diff=lfs merge=lfs -text
+1_2.png filter=lfs diff=lfs merge=lfs -text
+1_video.mp4 filter=lfs diff=lfs merge=lfs -text
+video.mp4 filter=lfs diff=lfs merge=lfs -text
+1_video_seed.mp4 filter=lfs diff=lfs merge=lfs -text
+2_video_seed.mp4 filter=lfs diff=lfs merge=lfs -text
+validition_v1/01/1.jpg filter=lfs diff=lfs merge=lfs -text
+validition_v1/01/2.jpg filter=lfs diff=lfs merge=lfs -text
+validition_v1/01/bg.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/02/1.jpg filter=lfs diff=lfs merge=lfs -text
+validition_v1/02/2.jpg filter=lfs diff=lfs merge=lfs -text
+validition_v1/02/bg.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/03/01.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/03/02.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/04/01.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/04/02.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/04/03.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/05/1.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/05/2.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/05/5.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/05/bg.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/05/video2.mp4 filter=lfs diff=lfs merge=lfs -text
+validition_v1/06/1.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/06/2.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/06/3.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/06/bg.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/07/1.jpg filter=lfs diff=lfs merge=lfs -text
+validition_v1/07/2.jpg filter=lfs diff=lfs merge=lfs -text
+validition_v1/07/bg.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/08/1.jpg filter=lfs diff=lfs merge=lfs -text
+validition_v1/08/2.jpg filter=lfs diff=lfs merge=lfs -text
+validition_v1/08/bg.png filter=lfs diff=lfs merge=lfs -text
+validition_v1/08/video1.mp4 filter=lfs diff=lfs merge=lfs -text
+validition_v1/08/video2.mp4 filter=lfs diff=lfs merge=lfs -text

LTX-2.3-Licon-MSR-V1.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4008d341c58791659a9b3840ac8d76f08d8941f82670dc2d3dad0f71876bbae5
+size 654443424

LTX-2.3_MSR_sample_workflow_V1.json

@@ -0,0 +1,1987 @@
+{
+  "id": "aaf551c7-d6b3-46a7-8996-8f20cd40a6bd",
+  "revision": 0,
+  "last_node_id": 79,
+  "last_link_id": 307,
+  "nodes": [
+    {
+      "id": 39,
+      "type": "LoadImage",
+      "pos": [
+        -1780,
+        -90
+      ],
+      "size": [
+        270,
+        320
+      ],
+      "flags": {},
+      "order": 0,
+      "mode": 4,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": []
+        },
+        {
+          "name": "MASK",
+          "type": "MASK",
+          "links": null
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LoadImage",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "杞藉叿pic1_0012.png",
+        "image"
+      ]
+    },
+    {
+      "id": 23,
+      "type": "LTXVConcatAVLatent",
+      "pos": [
+        1460,
+        -20
+      ],
+      "size": [
+        190,
+        50
+      ],
+      "flags": {},
+      "order": 23,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "video_latent",
+          "type": "LATENT",
+          "link": 282
+        },
+        {
+          "name": "audio_latent",
+          "type": "LATENT",
+          "link": 65
+        }
+      ],
+      "outputs": [
+        {
+          "name": "latent",
+          "type": "LATENT",
+          "links": [
+            281
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LTXVConcatAVLatent",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 27,
+      "type": "ManualSigmas",
+      "pos": [
+        1570,
+        -220
+      ],
+      "size": [
+        270,
+        60
+      ],
+      "flags": {},
+      "order": 1,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "SIGMAS",
+          "type": "SIGMAS",
+          "links": [
+            200
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.5.1",
+        "Node name for S&R": "ManualSigmas",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "version": "7.6.2",
+          "input_ue_unconnectable": {}
+        },
+        "enableTabs": false,
+        "tabWidth": 65,
+        "tabXOffset": 10,
+        "hasSecondTab": false,
+        "secondTabText": "Send Back",
+        "secondTabOffset": 80,
+        "secondTabWidth": 65
+      },
+      "widgets_values": [
+        "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+      ]
+    },
+    {
+      "id": 16,
+      "type": "SamplerCustomAdvanced",
+      "pos": [
+        1910,
+        -120
+      ],
+      "size": [
+        180,
+        110
+      ],
+      "flags": {},
+      "order": 24,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "noise",
+          "type": "NOISE",
+          "link": 50
+        },
+        {
+          "name": "guider",
+          "type": "GUIDER",
+          "link": 139
+        },
+        {
+          "name": "sampler",
+          "type": "SAMPLER",
+          "link": 52
+        },
+        {
+          "name": "sigmas",
+          "type": "SIGMAS",
+          "link": 200
+        },
+        {
+          "name": "latent_image",
+          "type": "LATENT",
+          "link": 281
+        }
+      ],
+      "outputs": [
+        {
+          "name": "output",
+          "type": "LATENT",
+          "links": [
+            66
+          ]
+        },
+        {
+          "name": "denoised_output",
+          "type": "LATENT",
+          "links": null
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "SamplerCustomAdvanced",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 24,
+      "type": "LTXVSeparateAVLatent",
+      "pos": [
+        2010,
+        -410
+      ],
+      "size": [
+        190,
+        50
+      ],
+      "flags": {},
+      "order": 25,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "av_latent",
+          "type": "LATENT",
+          "link": 66
+        }
+      ],
+      "outputs": [
+        {
+          "name": "video_latent",
+          "type": "LATENT",
+          "links": [
+            57
+          ]
+        },
+        {
+          "name": "audio_latent",
+          "type": "LATENT",
+          "links": [
+            67
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LTXVSeparateAVLatent",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 25,
+      "type": "LTXVAudioVAEDecode",
+      "pos": [
+        2320,
+        -400
+      ],
+      "size": [
+        160,
+        50
+      ],
+      "flags": {},
+      "order": 27,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "samples",
+          "type": "LATENT",
+          "link": 67
+        },
+        {
+          "label": "Audio VAE",
+          "name": "audio_vae",
+          "type": "VAE",
+          "link": 167
+        }
+      ],
+      "outputs": [
+        {
+          "name": "Audio",
+          "type": "AUDIO",
+          "links": [
+            61
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LTXVAudioVAEDecode",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 9,
+      "type": "LTXAddVideoICLoRAGuide",
+      "pos": [
+        1170,
+        -650
+      ],
+      "size": [
+        320,
+        290
+      ],
+      "flags": {},
+      "order": 21,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "link": 179
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "link": 180
+        },
+        {
+          "name": "vae",
+          "type": "VAE",
+          "link": 42
+        },
+        {
+          "name": "latent",
+          "type": "LATENT",
+          "link": 43
+        },
+        {
+          "name": "image",
+          "type": "IMAGE",
+          "link": 154
+        }
+      ],
+      "outputs": [
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "links": [
+            152,
+            222
+          ]
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "links": [
+            153,
+            223
+          ]
+        },
+        {
+          "name": "latent",
+          "type": "LATENT",
+          "links": [
+            282
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "ComfyUI-LTXVideo",
+        "ver": "229437c6b65796d6a7a63ae34be2bd5ba31fa543",
+        "Node name for S&R": "LTXAddVideoICLoRAGuide",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        0,
+        1,
+        1,
+        "center",
+        false,
+        256,
+        64
+      ]
+    },
+    {
+      "id": 37,
+      "type": "CFGGuider",
+      "pos": [
+        1620,
+        -510
+      ],
+      "size": [
+        270,
+        110
+      ],
+      "flags": {},
+      "order": 22,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "link": 304
+        },
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "link": 222
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "link": 223
+        }
+      ],
+      "outputs": [
+        {
+          "name": "GUIDER",
+          "type": "GUIDER",
+          "links": [
+            139
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.64",
+        "Node name for S&R": "CFGGuider",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "version": "7.6.2",
+          "input_ue_unconnectable": {}
+        },
+        "enableTabs": false,
+        "tabWidth": 65,
+        "tabXOffset": 10,
+        "hasSecondTab": false,
+        "secondTabText": "Send Back",
+        "secondTabOffset": 80,
+        "secondTabWidth": 65
+      },
+      "widgets_values": [
+        1
+      ]
+    },
+    {
+      "id": 3,
+      "type": "LowVRAMCheckpointLoader",
+      "pos": [
+        -780,
+        -60
+      ],
+      "size": [
+        350,
+        100
+      ],
+      "flags": {},
+      "order": 2,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "dependencies",
+          "shape": 7,
+          "type": "*",
+          "link": null
+        }
+      ],
+      "outputs": [
+        {
+          "name": "MODEL",
+          "type": "MODEL",
+          "links": [
+            45
+          ]
+        },
+        {
+          "name": "CLIP",
+          "type": "CLIP",
+          "links": null
+        },
+        {
+          "name": "VAE",
+          "type": "VAE",
+          "links": [
+            42,
+            97
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "LowVRAMCheckpointLoader",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "ltx-2.3-22b-distilled-1.1.safetensors"
+      ]
+    },
+    {
+      "id": 26,
+      "type": "LTXAVTextEncoderLoader",
+      "pos": [
+        -930,
+        260
+      ],
+      "size": [
+        420,
+        110
+      ],
+      "flags": {},
+      "order": 3,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "CLIP",
+          "type": "CLIP",
+          "links": [
+            292,
+            294
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.7.0",
+        "Node name for S&R": "LTXAVTextEncoderLoader",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "version": "7.6.2",
+          "input_ue_unconnectable": {}
+        },
+        "models": [
+          {
+            "name": "ltx-2.3-22b-dev.safetensors",
+            "url": "https://huggingface.co/Lightricks/LTX-2.3/resolve/main/ltx-2.3-22b-dev.safetensors",
+            "directory": "checkpoints"
+          },
+          {
+            "name": "gemma_3_12B_it_fp4_mixed.safetensors",
+            "url": "https://huggingface.co/Comfy-Org/ltx-2/resolve/main/split_files/text_encoders/gemma_3_12B_it_fp4_mixed.safetensors",
+            "directory": "text_encoders"
+          }
+        ],
+        "enableTabs": false,
+        "tabWidth": 65,
+        "tabXOffset": 10,
+        "hasSecondTab": false,
+        "secondTabText": "Send Back",
+        "secondTabOffset": 80,
+        "secondTabWidth": 65
+      },
+      "widgets_values": [
+        "gemma_3_12B_it.safetensors",
+        "ltx-2.3-22b-distilled-1.1.safetensors",
+        "default"
+      ]
+    },
+    {
+      "id": 20,
+      "type": "SaveVideo",
+      "pos": [
+        840,
+        -1560
+      ],
+      "size": [
+        750,
+        760
+      ],
+      "flags": {},
+      "order": 30,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "video",
+          "type": "VIDEO",
+          "link": 62
+        }
+      ],
+      "outputs": [],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "SaveVideo",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "LTX-2/MSR_verify_iclora_guide",
+        "auto",
+        "auto"
+      ]
+    },
+    {
+      "id": 28,
+      "type": "LiconMSR",
+      "pos": [
+        740,
+        -580
+      ],
+      "size": [
+        270,
+        190
+      ],
+      "flags": {},
+      "order": 18,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "1",
+          "shape": 7,
+          "type": "IMAGE",
+          "link": 240
+        },
+        {
+          "name": "2",
+          "shape": 7,
+          "type": "IMAGE",
+          "link": 241
+        },
+        {
+          "name": "3",
+          "shape": 7,
+          "type": "IMAGE",
+          "link": 242
+        },
+        {
+          "name": "4",
+          "shape": 7,
+          "type": "IMAGE",
+          "link": null
+        },
+        {
+          "name": "background",
+          "shape": 7,
+          "type": "IMAGE",
+          "link": 75
+        },
+        {
+          "name": "width",
+          "type": "INT",
+          "widget": {
+            "name": "width"
+          },
+          "link": 119
+        },
+        {
+          "name": "height",
+          "type": "INT",
+          "widget": {
+            "name": "height"
+          },
+          "link": 121
+        }
+      ],
+      "outputs": [
+        {
+          "name": "output",
+          "type": "IMAGE",
+          "links": [
+            154
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "LiconMSR",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        1920,
+        1280,
+        41
+      ]
+    },
+    {
+      "id": 29,
+      "type": "LoadImage",
+      "pos": [
+        440,
+        -1560
+      ],
+      "size": [
+        270,
+        320
+      ],
+      "flags": {},
+      "order": 4,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": [
+            240
+          ]
+        },
+        {
+          "name": "MASK",
+          "type": "MASK",
+          "links": null
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LoadImage",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "2.jpg",
+        "image"
+      ]
+    },
+    {
+      "id": 40,
+      "type": "LoadImage",
+      "pos": [
+        430,
+        -1130
+      ],
+      "size": [
+        270,
+        320
+      ],
+      "flags": {},
+      "order": 5,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": [
+            241
+          ]
+        },
+        {
+          "name": "MASK",
+          "type": "MASK",
+          "links": null
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LoadImage",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "1.jpg",
+        "image"
+      ]
+    },
+    {
+      "id": 19,
+      "type": "CreateVideo",
+      "pos": [
+        2590,
+        -820
+      ],
+      "size": [
+        270,
+        80
+      ],
+      "flags": {},
+      "order": 29,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "images",
+          "type": "IMAGE",
+          "link": 305
+        },
+        {
+          "name": "audio",
+          "shape": 7,
+          "type": "AUDIO",
+          "link": 61
+        }
+      ],
+      "outputs": [
+        {
+          "name": "VIDEO",
+          "type": "VIDEO",
+          "links": [
+            62
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "CreateVideo",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        50
+      ]
+    },
+    {
+      "id": 50,
+      "type": "INTConstant",
+      "pos": [
+        -340,
+        -560
+      ],
+      "size": [
+        210,
+        60
+      ],
+      "flags": {},
+      "order": 6,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "值",
+          "name": "value",
+          "type": "INT",
+          "widget": {
+            "name": "value"
+          },
+          "link": null
+        }
+      ],
+      "outputs": [
+        {
+          "label": "值",
+          "name": "value",
+          "type": "INT",
+          "links": [
+            230,
+            231
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfyui-kjnodes",
+        "ver": "c661baadd9683c0033cd2a6ad90157c6d099a6c2",
+        "Node name for S&R": "INTConstant",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.4.1"
+        },
+        "widget_ue_connectable": {
+          "value": true
+        }
+      },
+      "widgets_values": [
+        751
+      ],
+      "color": "#1b4669",
+      "bgcolor": "#29699c"
+    },
+    {
+      "id": 33,
+      "type": "LoadImage",
+      "pos": [
+        360,
+        -580
+      ],
+      "size": [
+        270,
+        320
+      ],
+      "flags": {},
+      "order": 7,
+      "mode": 4,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": [
+            242
+          ]
+        },
+        {
+          "name": "MASK",
+          "type": "MASK",
+          "links": null
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LoadImage",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "猪.png",
+        "image"
+      ]
+    },
+    {
+      "id": 43,
+      "type": "INTConstant",
+      "pos": [
+        -260,
+        -810
+      ],
+      "size": [
+        210,
+        60
+      ],
+      "flags": {},
+      "order": 8,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "值",
+          "name": "value",
+          "type": "INT",
+          "widget": {
+            "name": "value"
+          },
+          "link": null
+        }
+      ],
+      "outputs": [
+        {
+          "label": "值",
+          "name": "value",
+          "type": "INT",
+          "links": [
+            119,
+            122
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfyui-kjnodes",
+        "ver": "c661baadd9683c0033cd2a6ad90157c6d099a6c2",
+        "Node name for S&R": "INTConstant",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.4.1"
+        },
+        "widget_ue_connectable": {
+          "value": true
+        }
+      },
+      "widgets_values": [
+        1280
+      ],
+      "color": "#1b4669",
+      "bgcolor": "#29699c"
+    },
+    {
+      "id": 44,
+      "type": "INTConstant",
+      "pos": [
+        -220,
+        -680
+      ],
+      "size": [
+        210,
+        60
+      ],
+      "flags": {},
+      "order": 9,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "值",
+          "name": "value",
+          "type": "INT",
+          "widget": {
+            "name": "value"
+          },
+          "link": null
+        }
+      ],
+      "outputs": [
+        {
+          "label": "值",
+          "name": "value",
+          "type": "INT",
+          "links": [
+            120,
+            121
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfyui-kjnodes",
+        "ver": "c661baadd9683c0033cd2a6ad90157c6d099a6c2",
+        "Node name for S&R": "INTConstant",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.4.1"
+        },
+        "widget_ue_connectable": {
+          "value": true
+        }
+      },
+      "widgets_values": [
+        720
+      ],
+      "color": "#1b4669",
+      "bgcolor": "#29699c"
+    },
+    {
+      "id": 30,
+      "type": "LoadImage",
+      "pos": [
+        40,
+        -590
+      ],
+      "size": [
+        270,
+        320
+      ],
+      "flags": {},
+      "order": 10,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": [
+            75
+          ]
+        },
+        {
+          "name": "MASK",
+          "type": "MASK",
+          "links": null
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LoadImage",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "bg (5).png",
+        "image"
+      ]
+    },
+    {
+      "id": 10,
+      "type": "LTXICLoRALoaderModelOnly",
+      "pos": [
+        -390,
+        -80
+      ],
+      "size": [
+        390,
+        140
+      ],
+      "flags": {},
+      "order": 14,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "link": 45
+        }
+      ],
+      "outputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "links": [
+            304
+          ]
+        },
+        {
+          "name": "latent_downscale_factor",
+          "type": "FLOAT",
+          "links": null
+        }
+      ],
+      "properties": {
+        "cnr_id": "ComfyUI-LTXVideo",
+        "ver": "229437c6b65796d6a7a63ae34be2bd5ba31fa543",
+        "Node name for S&R": "LTXICLoRALoaderModelOnly",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "LTX-2.3\\LTX-2.3-Licon-MSR-V1.safetensors",
+        1
+      ]
+    },
+    {
+      "id": 17,
+      "type": "LTXVCropGuides",
+      "pos": [
+        1950,
+        -710
+      ],
+      "size": [
+        220,
+        70
+      ],
+      "flags": {},
+      "order": 26,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "link": 152
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "link": 153
+        },
+        {
+          "name": "latent",
+          "type": "LATENT",
+          "link": 57
+        }
+      ],
+      "outputs": [
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "links": []
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "links": []
+        },
+        {
+          "name": "latent",
+          "type": "LATENT",
+          "links": [
+            149
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LTXVCropGuides",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 38,
+      "type": "VAEDecode",
+      "pos": [
+        2240,
+        -760
+      ],
+      "size": [
+        240,
+        50
+      ],
+      "flags": {},
+      "order": 28,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "samples",
+          "type": "LATENT",
+          "link": 149
+        },
+        {
+          "name": "vae",
+          "type": "VAE",
+          "link": 97
+        }
+      ],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": [
+            305
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.75",
+        "Node name for S&R": "VAEDecode",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "version": "7.6.2",
+          "input_ue_unconnectable": {}
+        },
+        "enableTabs": false,
+        "tabWidth": 65,
+        "tabXOffset": 10,
+        "hasSecondTab": false,
+        "secondTabText": "Send Back",
+        "secondTabOffset": 80,
+        "secondTabWidth": 65
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 5,
+      "type": "CLIPTextEncode",
+      "pos": [
+        290,
+        -130
+      ],
+      "size": [
+        430,
+        280
+      ],
+      "flags": {},
+      "order": 15,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "clip",
+          "type": "CLIP",
+          "link": 292
+        }
+      ],
+      "outputs": [
+        {
+          "name": "CONDITIONING",
+          "type": "CONDITIONING",
+          "links": [
+            38
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "CLIPTextEncode",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "Realistic cinematic Chinese fantasy style, maintain strong reference consistency for both characters and the ancient village street. Do not change their faces, hairstyles, costumes, colors, accessories, fabric patterns, body proportions, or overall visual style.\n\n15-second one-shot video, no cuts, high-motion chase sequence. After the rain, an ancient Jiangnan village alley with white walls, black tiled roofs, wet stone pavement, warm glowing lanterns, misty mountains in the distance, and willow branches hanging in the foreground.\n\nThe camera starts behind the wet willow branches, then rushes forward rapidly into the narrow alley. The woman in the pale green ornate ancient dress suddenly runs across the frame from right to left, her long layered skirt, translucent shawl, jade ornaments, earrings, and hair ribbons flying dramatically in the wind. She glances back with a tense expression while running.\n\nThe man in the deep red embroidered robe with a black fur collar bursts out from the left-side alley entrance and immediately chases after her. His long hair, pearl earrings, wide sleeves, red robe, and black fur collar move violently with the speed. His footsteps hit the wet stone pavement, splashing water from puddles.\n\nThe camera follows them in one continuous fast tracking shot, weaving through the narrow alley. The woman dodges around a stone step, spins past a glowing lantern, and her green shawl sweeps across the frame. The man closes the distance, reaches out toward her, but she quickly turns sideways and slips past him. Their sleeves nearly touch, creating a strong flowing motion between red and green fabric.\n\nThe camera circles around them as they pass each other, then instantly swings behind the man as he turns and continues chasing her deeper into the alley. Lantern reflections stretch and break across the wet ground. Willow leaves, mist, robe hems, hair strands, earrings, and translucent fabric all move strongly with the wind and speed.\n\nIn the final seconds, the woman suddenly stops, spins around sharply, and her long green dress forms a wide circular arc on the wet stone pavement. The man rushes forward and stops just in front of her, his red sleeves and robe continuing to fly forward from inertia. The camera pushes in quickly from a low angle and ends on their intense face-to-face stare. Strong dynamic motion, fast chase, dramatic fabric movement, water splashes, camera rotation, rapid push-in, wind-blown hair and accessories. No cuts, no face change, no costume change, no style change.\n"
+      ]
+    },
+    {
+      "id": 6,
+      "type": "CLIPTextEncode",
+      "pos": [
+        310,
+        230
+      ],
+      "size": [
+        400,
+        200
+      ],
+      "flags": {},
+      "order": 16,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "clip",
+          "type": "CLIP",
+          "link": 294
+        }
+      ],
+      "outputs": [
+        {
+          "name": "CONDITIONING",
+          "type": "CONDITIONING",
+          "links": [
+            39
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "CLIPTextEncode",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "worst quality, blurry, jittery, distorted, inconsistent appearance，"
+      ]
+    },
+    {
+      "id": 21,
+      "type": "LTXVAudioVAELoader",
+      "pos": [
+        -420,
+        440
+      ],
+      "size": [
+        270,
+        60
+      ],
+      "flags": {},
+      "order": 11,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "Audio VAE",
+          "type": "VAE",
+          "links": [
+            63,
+            167
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LTXVAudioVAELoader",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "ltx-2.3-22b-distilled-1.1.safetensors"
+      ]
+    },
+    {
+      "id": 7,
+      "type": "LTXVConditioning",
+      "pos": [
+        890,
+        -120
+      ],
+      "size": [
+        270,
+        80
+      ],
+      "flags": {},
+      "order": 20,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "link": 38
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "link": 39
+        }
+      ],
+      "outputs": [
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "links": [
+            179
+          ]
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "links": [
+            180
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LTXVConditioning",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        50
+      ]
+    },
+    {
+      "id": 8,
+      "type": "EmptyLTXVLatentVideo",
+      "pos": [
+        830,
+        150
+      ],
+      "size": [
+        270,
+        130
+      ],
+      "flags": {},
+      "order": 17,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "width",
+          "type": "INT",
+          "widget": {
+            "name": "width"
+          },
+          "link": 122
+        },
+        {
+          "name": "height",
+          "type": "INT",
+          "widget": {
+            "name": "height"
+          },
+          "link": 120
+        },
+        {
+          "name": "length",
+          "type": "INT",
+          "widget": {
+            "name": "length"
+          },
+          "link": 230
+        }
+      ],
+      "outputs": [
+        {
+          "name": "LATENT",
+          "type": "LATENT",
+          "links": [
+            43
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "EmptyLTXVLatentVideo",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        1280,
+        1920,
+        145,
+        1
+      ]
+    },
+    {
+      "id": 15,
+      "type": "RandomNoise",
+      "pos": [
+        1440,
+        160
+      ],
+      "size": [
+        290,
+        90
+      ],
+      "flags": {},
+      "order": 12,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "NOISE",
+          "type": "NOISE",
+          "links": [
+            50
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "RandomNoise",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        97593527707150,
+        "randomize"
+      ]
+    },
+    {
+      "id": 13,
+      "type": "KSamplerSelect",
+      "pos": [
+        1850,
+        120
+      ],
+      "size": [
+        270,
+        60
+      ],
+      "flags": {},
+      "order": 13,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "SAMPLER",
+          "type": "SAMPLER",
+          "links": [
+            52
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "KSamplerSelect",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        "euler"
+      ]
+    },
+    {
+      "id": 22,
+      "type": "LTXVEmptyLatentAudio",
+      "pos": [
+        1140,
+        30
+      ],
+      "size": [
+        270,
+        110
+      ],
+      "flags": {},
+      "order": 19,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "Audio VAE",
+          "name": "audio_vae",
+          "type": "VAE",
+          "link": 63
+        },
+        {
+          "name": "frames_number",
+          "type": "INT",
+          "widget": {
+            "name": "frames_number"
+          },
+          "link": 231
+        }
+      ],
+      "outputs": [
+        {
+          "name": "Latent",
+          "type": "LATENT",
+          "links": [
+            65
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.20.1",
+        "Node name for S&R": "LTXVEmptyLatentAudio",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {},
+          "version": "7.8"
+        }
+      },
+      "widgets_values": [
+        241,
+        50,
+        1
+      ]
+    }
+  ],
+  "links": [
+    [
+      38,
+      5,
+      0,
+      7,
+      0,
+      "CONDITIONING"
+    ],
+    [
+      39,
+      6,
+      0,
+      7,
+      1,
+      "CONDITIONING"
+    ],
+    [
+      42,
+      3,
+      2,
+      9,
+      2,
+      "VAE"
+    ],
+    [
+      43,
+      8,
+      0,
+      9,
+      3,
+      "LATENT"
+    ],
+    [
+      45,
+      3,
+      0,
+      10,
+      0,
+      "MODEL"
+    ],
+    [
+      50,
+      15,
+      0,
+      16,
+      0,
+      "NOISE"
+    ],
+    [
+      52,
+      13,
+      0,
+      16,
+      2,
+      "SAMPLER"
+    ],
+    [
+      57,
+      24,
+      0,
+      17,
+      2,
+      "LATENT"
+    ],
+    [
+      61,
+      25,
+      0,
+      19,
+      1,
+      "AUDIO"
+    ],
+    [
+      62,
+      19,
+      0,
+      20,
+      0,
+      "VIDEO"
+    ],
+    [
+      63,
+      21,
+      0,
+      22,
+      0,
+      "VAE"
+    ],
+    [
+      65,
+      22,
+      0,
+      23,
+      1,
+      "LATENT"
+    ],
+    [
+      66,
+      16,
+      0,
+      24,
+      0,
+      "LATENT"
+    ],
+    [
+      67,
+      24,
+      1,
+      25,
+      0,
+      "LATENT"
+    ],
+    [
+      75,
+      30,
+      0,
+      28,
+      4,
+      "IMAGE"
+    ],
+    [
+      97,
+      3,
+      2,
+      38,
+      1,
+      "VAE"
+    ],
+    [
+      119,
+      43,
+      0,
+      28,
+      5,
+      "INT"
+    ],
+    [
+      120,
+      44,
+      0,
+      8,
+      1,
+      "INT"
+    ],
+    [
+      121,
+      44,
+      0,
+      28,
+      6,
+      "INT"
+    ],
+    [
+      122,
+      43,
+      0,
+      8,
+      0,
+      "INT"
+    ],
+    [
+      139,
+      37,
+      0,
+      16,
+      1,
+      "GUIDER"
+    ],
+    [
+      149,
+      17,
+      2,
+      38,
+      0,
+      "LATENT"
+    ],
+    [
+      152,
+      9,
+      0,
+      17,
+      0,
+      "CONDITIONING"
+    ],
+    [
+      153,
+      9,
+      1,
+      17,
+      1,
+      "CONDITIONING"
+    ],
+    [
+      154,
+      28,
+      0,
+      9,
+      4,
+      "IMAGE"
+    ],
+    [
+      167,
+      21,
+      0,
+      25,
+      1,
+      "VAE"
+    ],
+    [
+      179,
+      7,
+      0,
+      9,
+      0,
+      "CONDITIONING"
+    ],
+    [
+      180,
+      7,
+      1,
+      9,
+      1,
+      "CONDITIONING"
+    ],
+    [
+      200,
+      27,
+      0,
+      16,
+      3,
+      "SIGMAS"
+    ],
+    [
+      222,
+      9,
+      0,
+      37,
+      1,
+      "CONDITIONING"
+    ],
+    [
+      223,
+      9,
+      1,
+      37,
+      2,
+      "CONDITIONING"
+    ],
+    [
+      230,
+      50,
+      0,
+      8,
+      2,
+      "INT"
+    ],
+    [
+      231,
+      50,
+      0,
+      22,
+      1,
+      "INT"
+    ],
+    [
+      240,
+      29,
+      0,
+      28,
+      0,
+      "IMAGE"
+    ],
+    [
+      241,
+      40,
+      0,
+      28,
+      1,
+      "IMAGE"
+    ],
+    [
+      242,
+      33,
+      0,
+      28,
+      2,
+      "IMAGE"
+    ],
+    [
+      281,
+      23,
+      0,
+      16,
+      4,
+      "LATENT"
+    ],
+    [
+      282,
+      9,
+      2,
+      23,
+      0,
+      "LATENT"
+    ],
+    [
+      292,
+      26,
+      0,
+      5,
+      0,
+      "CLIP"
+    ],
+    [
+      294,
+      26,
+      0,
+      6,
+      0,
+      "CLIP"
+    ],
+    [
+      304,
+      10,
+      0,
+      37,
+      0,
+      "MODEL"
+    ],
+    [
+      305,
+      38,
+      0,
+      19,
+      0,
+      "IMAGE"
+    ]
+  ],
+  "groups": [],
+  "config": {},
+  "extra": {
+    "ue_links": [],
+    "ds": {
+      "scale": 0.45,
+      "offset": [
+        911.6670541914682,
+        1891.5079558841767
+      ]
+    },
+    "links_added_by_ue": [],
+    "frontendVersion": "1.42.15",
+    "VHS_latentpreview": true,
+    "VHS_latentpreviewrate": 0,
+    "VHS_MetadataImage": false,
+    "VHS_KeepIntermediate": false
+  },
+  "version": 0.4
+}

LTX2.3-Licon-MSR-test_version.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51121a7e9d9579734943db1ebf89df12592ef7e6cdda460eca4f9ab8ef989859
+size 805412808

README.md

@@ -0,0 +1,80 @@
+---
+license: apache-2.0
+tags:
+- video-generation
+- multi-reference
+- LTX-2.3
+base_model:
+- Lightricks/LTX-2.3
+github: https://github.com/liconstudio/ComfyUI-Licon-MSR
+library_name: diffusers
+---
+
+## Overview
+
+This model implements a novel approach to multi-reference video generation using **Multiple Subject Reference (MSR)**. Instead of introducing additional encoder branches or fusion modules, we transform multiple static reference images into a pseudo-video sequence that shares the same representation space as the target video.
+
+## Usage
+
+This LoRA requires the **[ComfyUI-Licon-MSR](https://github.com/liconstudio/ComfyUI-Licon-MSR)** plugin for ComfyUI. A sample workflow is included in the model files for easy testing and experimentation.
+
+## Key Features
+
+### Multi-Reference Visual Memory
+
+- **Token-level reference preservation**: Multiple reference images are encoded as video latents, preserving fine-grained visual information at token level rather than compressing into a single embedding
+- **Native self-attention retrieval**: The target video tokens directly access reference tokens through the model's existing self-attention mechanism—no new architectural components needed
+- **In-context conditioning**: References serve as "visual memory" within the main token sequence, not as external conditioning inputs
+
+### Flexible Reference Composition
+
+- **2 to 5 reference images**: Supports varying numbers of reference inputs with increasing complexity
+- **Complementary semantic roles**: Each reference image can carry different information:
+  - Subject identity
+  - Object/prop details
+  - Scene/background
+  - Local textures
+  - Multiple viewpoints
+
+
+## What It Can Do
+
+### Identity Preservation Across References
+
+Generate videos where multiple reference identities are simultaneously preserved:
+- Multiple characters from different reference images
+- Character + object combinations
+- Object + scene compositions
+
+### Relation-Based Composition
+
+Beyond mere identity preservation, the model can compose references based on textual relation descriptions:
+- Action interactions (handing, picking up, pushing)
+- Spatial relationships (left-right, foreground-background)
+- Temporal event structures (start → process → result)
+
+### Cross-Reference Attribute Selection
+
+The model learns to selectively retrieve attributes from different references:
+- Face from reference A, clothing from reference B
+- Object identity from one reference, pose/position from another
+- Background elements from scene references
+
+## Usage Tips (V1 Version)
+
+- **Prompt description**: Requires concise but accurate description of reference images. Over-description or under-description both lead to consistency degradation
+- **High-motion scenes**: 50fps recommended to ensure smooth motion coherence
+- **Generation reliability**: Typically requires 2-3 sampling runs to achieve accurate results
+
+
+## Results Showcase
+
+### V1 Version
+
+| Reference Images | Generated Video |
+|:---:|:---:|
+| <img src="validition_v1/01/1.jpg" width="80"> <img src="validition_v1/01/2.jpg" width="80"> <img src="validition_v1/01/bg.png" width="80"> | [▶ Play](validition_v1/01/video.mp4) |
+| <img src="validition_v1/07/1.jpg" width="80"> <img src="validition_v1/07/2.jpg" width="80"> <img src="validition_v1/07/bg.png" width="80"> | [▶ Play](validition_v1/07/video.mp4) |
+| <img src="validition_v1/05/1.png" width="70"> <img src="validition_v1/05/2.png" width="70"> <img src="validition_v1/05/5.png" width="70"> <img src="validition_v1/05/bg.png" width="70"> | [▶ Play](validition_v1/05/video.mp4) |
+
+---

validation/1_video.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64c7786527a8a7ff8081893aa6d468ebeaee425ae4e25104d210c6d33a17fb88
+size 3219089

validation/1_video_seed.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5edb2743137e8cac51c8a35d35fcb065f8c86ff12aaf8ed94a70f9a906d6211e
+size 2117685

validation/2_video.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96b2e6dd2a278ff4ce5cf51479019d679f4e19b002b8018de1a7badb103a9be2
+size 3424939

validation/2_video_seed.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e46153d2e4a654862024befa99fa78b995b6e12f15fe12df1c739f37acf8919
+size 6101580

validition_v1/01/prompt.txt

@@ -0,0 +1,7 @@
+参考图1：RN，男人，深棕色长发半束着，佩戴黑色龙形发簪与珍珠长流苏发带，身着酒红色暗纹广袖长袍，领口袖口绣精致金色凤凰纹样，外搭厚实黑色狐狸毛大氅，腰间系嵌绿玉的金纹黑色腰带，面容俊朗，仪态高贵。
+参考图2：WV，女人，乌黑长发盘成精致发髻，佩戴绿玉白花发饰与同色系水滴流苏耳坠，身着暗绿色抹胸鱼尾长裙，裙身绣立体金色花卉纹样，长裙的腰间镂空露出两侧腰部，露肩露背。外搭同色系飘逸透明广袖披帛，颈间佩戴绿玉平安扣项链，面容温婉清丽，眉眼柔和灵动。
+参考图3（场景）：烟雨朦胧的江南小镇，石板小路被雨水打湿形成镜面倒影，两侧是白墙黑瓦的建筑，黑瓦上挂着晶莹的雨珠，屋檐下悬挂着暖黄色的灯笼，街道中央有一棵枝繁叶茂的垂柳，远处是云雾缭绕的青山，整体氛围清冷诗意，充满江南水乡的韵味。
+石板小路上，WV迎面走来，双手轻轻提着裙摆，衣裙飘飘。WV走近，表情从冷淡转为微笑。
+RN迎面站着，也微微一笑，说：“这么巧，没想到会在这里遇见你”。
+WV和RN两人在石板小路上面对面站着，WV微微点头，脸颊微红，嘴角扬起浅浅的笑意，声音甜美说：“我也没想到，一起走走吧。”
+镜头拉远至中景，展示WV和RN两人的侧面全身视角。RN侧身做出邀请的手势，WV点头回应，两人并肩向街道深处走去。镜头缓缓跟拍，逐渐绕到两人身后，展示背面视角，看着他们的身影逐渐消失在烟雨朦胧的石板小路尽头，

validition_v1/01/video.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bfb5ceea78c3514d4b51428193e5b973194afd8090ec424e308e45cdc4ebd3a
+size 4683490

validition_v1/02/prompt.txt

@@ -0,0 +1,5 @@
+In a misty ancient Chinese water-town alley after rain, white plaster walls and black tiled roofs line the wet stone path, distant green mountains fade into soft fog, willow branches hang in the foreground, and warm lanterns glow under the eaves, reflecting on the rain-soaked ground.
+A beautiful young woman in an elegant pale jade-green fantasy hanfu stands on the left side of the alley. She has delicate calm facial features, long black hair styled in an ancient updo, jade-green hair ornaments, dangling earrings, and a green jade necklace. Her dress has translucent flowing shawls, silver floral embroidery, and layered gauze fabric, giving her a graceful and restrained presence.
+A handsome young man in a deep crimson ancient robe stands on the right side. He has sharp refined facial features, long black hair half tied back with hanging tassel ornaments, a black fur-collar cloak, and golden embroidered patterns on his red robe, giving him a composed and noble aura.
+
+A 10-second cinematic two-person dialogue scene: the woman first looks down at her reflection in the puddle, then slowly raises her eyes toward the man and speaks softly. The man remains silent for a moment, his expression subtly changes, then he steps half a step forward and replies in a low voice. They keep a restrained distance, like two people meeting again after a long separation, speaking with suppressed emotion in the rainy alley. The camera starts with a medium two-shot, slowly dollies forward with a slight circular movement around them, and ends on a half-body side-angle shot of the two characters looking at each other. Damp cool atmosphere, warm lantern light on their faces and clothing, sleeves and translucent fabric gently moving in the breeze, subtle hair movement, natural lip movement, delicate expressions, cinematic ancient costume drama style, realistic lighting, shallow depth of field, soft mist, high quality, stable character consistency.

validition_v1/02/video.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a04a497e4f73eb54e04f02a8ddb049587eb010edaff0883efcfbbebc07d5871a
+size 4836159

validition_v1/03/prompt.txt

@@ -0,0 +1,5 @@
+男人：
+人物形象：男人留着利落的黑色短发，佩戴一副细框金属眼镜，面容清秀斯文，透着一股儒雅的精英气质。身穿一件深炭灰色或黑色的长款毛呢大衣。大衣剪裁修身。
+提示词：
+【@男人】说话，台词内容：牧宅的规矩很简单，护住这几个人，活过五天。但没人告诉你，真正的鬼，往往就在你身后……或者，就是你身边的人。 时长：15秒 音色：年轻男性，普通话，25-30岁。 表演要求：低沉、压抑。语速微慢，每句话之间停顿约0.4秒。
+最后一句“或者，就是你身边的人”保持平直但紧张的收束感。不要环境音，只要人声。

validition_v1/03/video.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc885f1b5341c6288bce31f3a9565a677a0b2019fff09a9d0e08657f6798489b
+size 1665155

validition_v1/04/prompt.txt

@@ -0,0 +1,4 @@
+女人：一头波浪状的金色长发，皮肤白皙，五官精致立体。身穿一件质感极佳的银色缎面礼服，设计为露肩款式，领口呈V字形。
+男人：男子留着向后梳理的金色短发，面部轮廓硬朗，胡茬清晰可见，展现出成熟男性的沧桑感。
+房间：极具未来感和舒适氛围的高科技居住舱或微型公寓内部。整体设计融合了极简主义、赛博朋克元素与温馨的居家感。
+房间内。男人眼神温柔，迈前半步身体微晃，抬臂环住女人，手掌贴女人后背手指轻颤，女人身体紧绷后放松，脸靠肩头轻蹭，闭眼睫毛轻颤，手指蜷缩轻抓他衣侧，肩膀轻颤，男人手臂轻收又放松，女人深埋胸口闷声轻泣，男人下巴抵女人发顶闭眼，屋外灯光勾勒霓虹轮廓，镜头推进再环绕再横移，音效衣物摩擦声轻泣声心跳声 【00-05】镜头推进，男人迈前半步贴近，抬臂环住女人，手掌贴女人后背手指轻颤。 【05-10】镜头环绕，女人脸靠肩头轻蹭，手指抓他衣侧，肩膀轻颤。 【10-15】横移，女人深埋胸口闷声轻泣，男人下巴抵女人发顶闭眼。

validition_v1/04/video.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc08a7193ce6d51258d358c4f2934902fdf3f82e8d328dc181abf8f1f305e08d
+size 4285574

validition_v1/05/prompt.txt

@@ -0,0 +1 @@
+Under a clear blue sky, a bright outdoor racing track surrounded by green hills and lush vegetation. A glossy, vibrant red cartoon convertible sports car with white racing stripes. The video starts with close-up shots from the front, side, and rear angles of the car. The vehicle launches from a standstill with a roaring engine, departing from the starting line, accelerating forward and drifting steadily around the bend. Dynamic follow camera with smooth view transitions. Soft natural sunlight and 3D cartoon lighting effects, vivid and lively colors, immersive racing scene, consistent car design throughout.

validition_v1/05/video.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0dd64ec7787d7a9dd30610e706ca1edf3efb4a5d443aca94afae8313de3e015
+size 4802158

validition_v1/05/video2.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f020b08f2cd671ad19c75305b9888bb6a1c1059486659d02239e2b54ec121d1
+size 5555708

validition_v1/06/prompt.txt

@@ -0,0 +1,18 @@
+Farmer:
+An elderly farmer wearing a wide-brimmed straw hat and a yellow shirt, with a kind and gentle smile.
+
+Goat:
+A gentle-looking white goat wearing a brown collar with a golden bell. It has curved horns and calm, docile eyes.
+
+Pig:
+A round and adorable pink pig with an upward-curled tail, a cute and slightly silly expression, smooth delicate skin, and short but sturdy legs.
+
+Farm:
+A green grassy farm surrounded by wooden fences. On the right side stands a small shed with a blue metal roof. In the distance are golden wheat fields under a blue sky with white clouds. Warm sunlight falls across the scene, creating soft and gentle lighting and shadows.
+
+Prompt:
+Anime style, keep the original visual style unchanged, maintain strong reference consistency for the farmer, goat, pig, and farm environment. Fixed wide-angle shot, a bright and sunny countryside farm. On the left side of the frame is a wooden goat pen, and on the right side is a blue pig shed. An elderly farmer wearing a straw hat stands in the center with a kind, gentle smile. A white goat with a small bell around its neck is inside the goat pen, softly bleating. In the pig shed on the right, a pink little pig is sleeping.
+
+The farmer slowly walks toward the goat pen on the left, reaches through the fence, and gently strokes the goat’s head. The goat affectionately rubs against him, and the bell makes a soft, delicate sound. Then the farmer turns around and slowly walks toward the pig shed on the right. He bends down slightly and gently pats the pig’s head. The pig wakes up, lifts its head, sees the farmer, stands up, walks out of the shed, circles around the farmer once, and rubs its head against the farmer’s leg.
+
+A light breeze blows softly, making the grass and leaves sway slightly. The sunlight and shadows shift slowly and naturally. Low motion, with only subtle breathing-like movements and gentle natural details. Maintain the anime style throughout, do not change the character designs, proportions, colors, clothing, animal features, farm layout, or overall visual style from the references.

validition_v1/06/video.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6648e77d89b299c7e3aed0ab0711b4fc813e184f821ec9bbb2e321dd4643907d
+size 4827405