added multimodal mode

app.py

@@ -594,7 +594,7 @@ def build_app() -> gr.Blocks:
         # Each mode button injects a tiny on-click that rewrites #aio-mode-tag
         # and (on mobile) auto-collapses the drawer.
         _MODE_TAG_BY_NAME = {
-            "t2v": "T2V", "a2v": "A2V", "i2v": "I2V",
+            "t2v": "T2V", "a2v": "A2V", "i2v": "I2V", "multimodal": "MIAV",
             "lipsync": "LIPSYNC", "keyframe": "KEY", "style": "STYLE",
         }
         for name, btn in mode_buttons.items():
@@ -715,6 +715,9 @@ def _render_one_mode(name: str) -> dict:
                 handles["image"] = gr.Image(label="Source image", type="filepath")
             elif name == "a2v":
                 handles["audio"] = gr.Audio(label="Source audio", type="filepath")
+            elif name == "multimodal":
+                handles["image"] = gr.Image(label="Initial image", type="filepath")
+                handles["audio"] = gr.Audio(label="Driving audio", type="filepath")
             elif name == "lipsync":
                 handles["image"] = gr.Image(label="Portrait", type="filepath")
                 handles["audio"] = gr.Audio(label="Speech audio", type="filepath")
@@ -1064,6 +1067,8 @@ def _input_keys_for_mode(mode_name: str, h: dict) -> list[str]:
         base.append("image")
     elif mode_name == "a2v":
         base.append("audio")
+    elif mode_name == "multimodal":
+        base.extend(["image", "audio"])
     elif mode_name == "lipsync":
         base.extend(["image", "audio"])
     elif mode_name == "keyframe":
@@ -1088,6 +1093,8 @@ def _collect_inputs_for_mode(mode_name: str, h: dict) -> list:
         base.append(h["image"])
     elif mode_name == "a2v":
         base.append(h["audio"])
+    elif mode_name == "multimodal":
+        base.extend([h["image"], h["audio"]])
     elif mode_name == "lipsync":
         base.extend([h["image"], h["audio"]])
     elif mode_name == "keyframe":

backend.py

@@ -75,6 +75,7 @@ _BASE_DURATION_S: dict[str, int] = {
     "t2v": 90,
     "i2v": 90,
     "a2v": 120,
+    "multimodal": 240,
     "lipsync": 240,   # extra: audio encoder + audio VAE + extra LoRAs
     "keyframe": 180,
     "style": 360,     # extra: preprocessor (canny/dwpose/depth) + IC-LoRAs

modes.py

@@ -66,6 +66,7 @@ SEED_NODE_BY_MODE: dict[str, str] = {
     "a2v": "463:5540",
     "i2v": "209:5541",
     "lipsync": "521:5542",
+    "multimodal": "521:5542",
     "keyframe": "670:5543",
     "style": "5364:5545",
 }
@@ -115,6 +116,13 @@ def _lipsync_parameterize(inp: dict[str, Any]) -> list[Patch]:
     ]
 
 
+def _multimodal_parameterize(inp: dict[str, Any]) -> list[Patch]:
+    return _shared_patches(inp, "multimodal") + [
+        (NODE_IMAGE_1, "image", inp["image"]),
+        (NODE_AUDIO, "audio", inp["audio"]),
+    ]
+
+
 def _keyframe_parameterize(inp: dict[str, Any]) -> list[Patch]:
     return _shared_patches(inp, "keyframe") + [
         (NODE_IMAGE_1, "image", inp["first_frame"]),
@@ -157,6 +165,7 @@ _A2V_STAGES = [
 ]
 
 _LIPSYNC_STAGES = list(_A2V_STAGES)
+_MULTIMODAL_STAGES = list(_A2V_STAGES)
 _KEYFRAME_STAGES = [
     Stage("Encode prompt", 5),
     Stage("Encode keyframes", 5),
@@ -201,6 +210,13 @@ MODE_REGISTRY["lipsync"] = Mode(
     parameterize_fn=_lipsync_parameterize,
     stage_map=_LIPSYNC_STAGES,
 )
+MODE_REGISTRY["multimodal"] = Mode(
+    name="multimodal",
+    label="Text + Image + Audio",
+    icon="🎭",
+    parameterize_fn=_multimodal_parameterize,
+    stage_map=_MULTIMODAL_STAGES,
+)
 MODE_REGISTRY["keyframe"] = Mode(
     name="keyframe",
     label="Keyframe → Video",

tests/conftest.py

@@ -103,6 +103,15 @@ def canonical_inputs() -> dict[str, dict[str, Any]]:
             "fps": 24,
             "seed": 42,
         },
+        "multimodal": {
+            "prompt": "a cinematic multimodal generation",
+            "image": "/tmp/portrait.png",
+            "audio": "/tmp/speech.wav",
+            "preset": "balanced",
+            "frames": 81,
+            "fps": 24,
+            "seed": 42,
+        },
         "keyframe": {
             "prompt": "smooth transition between the two frames",
             "first_frame": "/tmp/start.png",

tests/test_modes.py

@@ -49,7 +49,7 @@ def test_t2v_and_i2v_in_registry():
     assert "i2v" in modes.MODE_REGISTRY
 
 
-@pytest.mark.parametrize("mode_name", ["a2v", "lipsync", "keyframe", "style"])
+@pytest.mark.parametrize("mode_name", ["a2v", "lipsync", "multimodal", "keyframe", "style"])
 def test_remaining_modes_parameterize_validates(mode_name, canonical_inputs):
     inputs = canonical_inputs[mode_name]
     mode = modes.MODE_REGISTRY[mode_name]
@@ -86,13 +86,14 @@ def test_style_parameterize_passes_input_video(canonical_inputs):
     assert canonical_inputs["style"]["input_video"] in [p[2] for p in patches]
 
 
-def test_mode_registry_has_all_six_keys():
-    """All six modes are in the registry now."""
+def test_mode_registry_has_all_keys():
+    """All seven modes are in the registry now."""
     assert set(modes.MODE_REGISTRY.keys()) == {
         "t2v",
         "a2v",
         "i2v",
         "lipsync",
+        "multimodal",
         "keyframe",
         "style",
     }

ui.py

@@ -80,6 +80,7 @@ CAMERA_LORAS: list[str] = [
 IC_LORAS_BY_MODE: dict[str, list[str]] = {
     "t2v": [],
     "a2v": [],
+    "multimodal": ["union", "pose-control"],
     "i2v": ["union", "pose-control"],
     "lipsync": ["pose-control"],
     "keyframe": ["union"],

workflow.py

@@ -14,7 +14,7 @@ from typing import Any
 
 WORKFLOWS_DIR = pathlib.Path(__file__).parent / "workflows"
 
-VALID_MODES: tuple[str, ...] = ("t2v", "a2v", "i2v", "lipsync", "keyframe", "style")
+VALID_MODES: tuple[str, ...] = ("t2v", "a2v", "i2v", "lipsync", "keyframe", "style", "multimodal")
 
 
 def load_template(mode: str) -> dict[str, Any]:

workflows/multimodal.json

@@ -0,0 +1,993 @@
+{
+  "101": {
+    "inputs": {
+      "model_name": "ltx-2.3-spatial-upscaler-x2-1.0.safetensors"
+    },
+    "class_type": "LatentUpscaleModelLoader",
+    "_meta": {
+      "title": "Load Latent Upscale Model"
+    }
+  },
+  "146": {
+    "inputs": {
+      "clip_name1": "gemma_3_12B_it_fp4_mixed.safetensors",
+      "clip_name2": "ltx-2.3_text_projection_bf16.safetensors",
+      "type": "ltxv",
+      "device": "default"
+    },
+    "class_type": "DualCLIPLoader",
+    "_meta": {
+      "title": "DualCLIPLoader"
+    }
+  },
+  "149": {
+    "inputs": {
+      "image": "influencer_mic_hd.png"
+    },
+    "class_type": "LoadImage",
+    "_meta": {
+      "title": "Load Image1"
+    }
+  },
+  "174": {
+    "inputs": {
+      "vae_name": "LTX23_video_vae_bf16.safetensors",
+      "device": "main_device",
+      "weight_dtype": "bf16"
+    },
+    "class_type": "VAELoaderKJ",
+    "_meta": {
+      "title": "VAELoader Video"
+    }
+  },
+  "175": {
+    "inputs": {
+      "vae_name": "LTX23_audio_vae_bf16.safetensors",
+      "device": "main_device",
+      "weight_dtype": "bf16"
+    },
+    "class_type": "VAELoaderKJ",
+    "_meta": {
+      "title": "VAELoader Audio"
+    }
+  },
+  "188": {
+    "inputs": {
+      "frame_rate": [
+        "5446",
+        0
+      ],
+      "loop_count": 0,
+      "filename_prefix": "LTX2.3/Video",
+      "format": "video/h264-mp4",
+      "pix_fmt": "yuv420p",
+      "crf": 8,
+      "save_metadata": false,
+      "trim_to_audio": false,
+      "pingpong": false,
+      "save_output": true,
+      "images": [
+        "217",
+        0
+      ],
+      "audio": [
+        "218",
+        0
+      ]
+    },
+    "class_type": "VHS_VideoCombine",
+    "_meta": {
+      "title": "Video Combine 🎥🅥🅗🅢"
+    }
+  },
+  "196": {
+    "inputs": {
+      "Xi": 7,
+      "Xf": 7,
+      "isfloatX": 0
+    },
+    "class_type": "mxSlider",
+    "_meta": {
+      "title": "Clip Length ( in seconds )"
+    }
+  },
+  "211": {
+    "inputs": {
+      "PowerLoraLoaderHeaderWidget": {
+        "type": "PowerLoraLoaderHeaderWidget"
+      },
+      "lora_1": {
+        "on": true,
+        "lora": "ltx-2.3-22b-distilled-lora-dynamic_fro09_avg_rank_105_bf16.safetensors",
+        "strength": 0.6
+      },
+      "lora_2": {
+        "on": false,
+        "lora": "ltx-2.3-22b-distilled-lora-384.safetensors",
+        "strength": 1
+      },
+      "lora_3": {
+        "on": false,
+        "lora": "ltx-2-19b-lora-camera-control-dolly-left.safetensors",
+        "strength": 1
+      },
+      "lora_4": {
+        "on": false,
+        "lora": "ltx-2-19b-lora-camera-control-dolly-right.safetensors",
+        "strength": 1
+      },
+      "lora_5": {
+        "on": true,
+        "lora": "ltx-2-19b-lora-camera-control-dolly-in.safetensors",
+        "strength": 1
+      },
+      "lora_6": {
+        "on": false,
+        "lora": "ltx-2-19b-lora-camera-control-dolly-out.safetensors",
+        "strength": 1
+      },
+      "lora_7": {
+        "on": false,
+        "lora": "ltx-2-19b-lora-camera-control-jib-up.safetensors",
+        "strength": 1
+      },
+      "lora_8": {
+        "on": false,
+        "lora": "ltx-2-19b-lora-camera-control-jib-down.safetensors",
+        "strength": 0.6
+      },
+      "➕ Add Lora": "",
+      "model": [
+        "366",
+        0
+      ],
+      "clip": [
+        "146",
+        0
+      ]
+    },
+    "class_type": "Power Lora Loader (rgthree)",
+    "_meta": {
+      "title": "Power Lora Loader (rgthree)"
+    }
+  },
+  "217": {
+    "inputs": {
+      "any_04": [
+        "521:522",
+        0
+      ]
+    },
+    "class_type": "Any Switch (rgthree)",
+    "_meta": {
+      "title": "Any Switch image"
+    }
+  },
+  "218": {
+    "inputs": {
+      "any_04": [
+        "5400",
+        0
+      ]
+    },
+    "class_type": "Any Switch (rgthree)",
+    "_meta": {
+      "title": "Any Switch audio"
+    }
+  },
+  "366": {
+    "inputs": {
+      "unet_name": "ltx-2-3-22b-dev-Q4_K_M.gguf"
+    },
+    "class_type": "UnetLoaderGGUF",
+    "_meta": {
+      "title": "Unet Loader (GGUF) for Low V-Ram 8G-12G"
+    }
+  },
+  "591": {
+    "inputs": {
+      "vae_name": "taeltx2_3.safetensors"
+    },
+    "class_type": "VAELoader",
+    "_meta": {
+      "title": "Load VAE"
+    }
+  },
+  "700": {
+    "inputs": {
+      "chunks": 4,
+      "dim_threshold": 4096,
+      "model": [
+        "211",
+        0
+      ]
+    },
+    "class_type": "LTXVChunkFeedForward",
+    "_meta": {
+      "title": "LTXV Chunk FeedForward"
+    }
+  },
+  "5376": {
+    "inputs": {
+      "lora_name": "ltx-2.3-22b-ic-lora-union-control-ref0.5.safetensors",
+      "strength_model": 1,
+      "model": [
+        "211",
+        0
+      ]
+    },
+    "class_type": "LTXICLoRALoaderModelOnly",
+    "_meta": {
+      "title": "🅛🅣🅧 IC-LoRA Loader Model Only"
+    }
+  },
+  "5382": {
+    "inputs": {
+      "value": 1280
+    },
+    "class_type": "INTConstant",
+    "_meta": {
+      "title": "Height"
+    }
+  },
+  "5383": {
+    "inputs": {
+      "value": 720
+    },
+    "class_type": "INTConstant",
+    "_meta": {
+      "title": "Width"
+    }
+  },
+  "5387": {
+    "inputs": {
+      "expression": "a*b+1",
+      "a": [
+        "196",
+        0
+      ],
+      "b": [
+        "5445",
+        0
+      ]
+    },
+    "class_type": "MathExpression|pysssss",
+    "_meta": {
+      "title": "Math Expression 1🐍"
+    }
+  },
+  "5392": {
+    "inputs": {
+      "chunks": 4,
+      "dim_threshold": 4096,
+      "model": [
+        "5376",
+        0
+      ]
+    },
+    "class_type": "LTXVChunkFeedForward",
+    "_meta": {
+      "title": "LTXV Chunk FeedForward"
+    }
+  },
+  "5400": {
+    "inputs": {
+      "audio": "beauty_pagent_dialogue.mp3",
+      "start_time": 0,
+      "duration": [
+        "5442",
+        0
+      ]
+    },
+    "class_type": "VHS_LoadAudioUpload",
+    "_meta": {
+      "title": "Load Audio (Upload)🎥🅥🅗🅢"
+    }
+  },
+  "5401": {
+    "inputs": {
+      "audioUI": "",
+      "audio": [
+        "5400",
+        0
+      ]
+    },
+    "class_type": "PreviewAudio",
+    "_meta": {
+      "title": "Preview Audio"
+    }
+  },
+  "5429": {
+    "inputs": {
+      "resize_type": "scale dimensions",
+      "resize_type.width": [
+        "5383",
+        0
+      ],
+      "resize_type.height": [
+        "5382",
+        0
+      ],
+      "resize_type.crop": "center",
+      "scale_method": "lanczos",
+      "input": [
+        "149",
+        0
+      ]
+    },
+    "class_type": "ResizeImageMaskNode",
+    "_meta": {
+      "title": "Resize Image/Mask"
+    }
+  },
+  "5434": {
+    "inputs": {
+      "resize_type": "scale dimensions",
+      "resize_type.width": [
+        "5383",
+        0
+      ],
+      "resize_type.height": [
+        "5382",
+        0
+      ],
+      "resize_type.crop": "center",
+      "scale_method": "lanczos",
+      "input": [
+        "5437",
+        0
+      ]
+    },
+    "class_type": "ResizeImageMaskNode",
+    "_meta": {
+      "title": "Resize Image/Mask"
+    }
+  },
+  "5437": {
+    "inputs": {
+      "image": "5.FLF2.png"
+    },
+    "class_type": "LoadImage",
+    "_meta": {
+      "title": "Load Image2"
+    }
+  },
+  "5442": {
+    "inputs": {
+      "a": [
+        "196",
+        0
+      ]
+    },
+    "class_type": "CM_IntToFloat",
+    "_meta": {
+      "title": "IntToFloat"
+    }
+  },
+  "5444": {
+    "inputs": {
+      "video": "Sway Dance Lesson 31s.mp4",
+      "force_rate": [
+        "5446",
+        0
+      ],
+      "custom_width": 0,
+      "custom_height": 0,
+      "frame_load_cap": [
+        "5387",
+        0
+      ],
+      "skip_first_frames": 266,
+      "select_every_nth": 1,
+      "format": "AnimateDiff"
+    },
+    "class_type": "VHS_LoadVideo",
+    "_meta": {
+      "title": "Load Video (Upload) 🎥🅥🅗🅢"
+    }
+  },
+  "5445": {
+    "inputs": {
+      "value": 25
+    },
+    "class_type": "INTConstant",
+    "_meta": {
+      "title": "FPS"
+    }
+  },
+  "5446": {
+    "inputs": {
+      "a": [
+        "5445",
+        0
+      ]
+    },
+    "class_type": "CM_IntToFloat",
+    "_meta": {
+      "title": "IntToFloat"
+    }
+  },
+  "5458": {
+    "inputs": {
+      "resize_type": "scale dimensions",
+      "resize_type.width": [
+        "5383",
+        0
+      ],
+      "resize_type.height": [
+        "5382",
+        0
+      ],
+      "resize_type.crop": "center",
+      "scale_method": "lanczos",
+      "input": [
+        "5444",
+        0
+      ]
+    },
+    "class_type": "ResizeImageMaskNode",
+    "_meta": {
+      "title": "Resize Image/Mask"
+    }
+  },
+  "5536": {
+    "inputs": {
+      "text": "A woman talking, her face and gestures demonstrate the conversation content. The camera slowly dolly in.",
+      "clip": [
+        "146",
+        0
+      ]
+    },
+    "class_type": "CLIPTextEncode",
+    "_meta": {
+      "title": "CLIP Text Encode (Prompt) positive"
+    }
+  },
+  "5537": {
+    "inputs": {
+      "text": "blurry, low quality, still frame, frames, watermark, overlay, titles, has blurbox, has subtitles",
+      "clip": [
+        "146",
+        0
+      ]
+    },
+    "class_type": "CLIPTextEncode",
+    "_meta": {
+      "title": "CLIP Text Encode (Prompt) negative"
+    }
+  },
+  "5560": {
+    "inputs": {
+      "unet_name": "ltx-2.3-22b-dev_transformer_only_fp8_scaled.safetensors",
+      "weight_dtype": "default"
+    },
+    "class_type": "UNETLoader",
+    "_meta": {
+      "title": "Diffusion Model Loader for V-ram 16G up"
+    }
+  },
+  "521:465": {
+    "inputs": {
+      "sigmas": "1., 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+    },
+    "class_type": "ManualSigmas",
+    "_meta": {
+      "title": "ManualSigmas"
+    }
+  },
+  "521:469": {
+    "inputs": {
+      "value": 0,
+      "width": [
+        "521:473",
+        0
+      ],
+      "height": [
+        "521:473",
+        1
+      ]
+    },
+    "class_type": "SolidMask",
+    "_meta": {
+      "title": "SolidMask"
+    }
+  },
+  "521:471": {
+    "inputs": {
+      "width": [
+        "521:473",
+        0
+      ],
+      "height": [
+        "521:473",
+        1
+      ],
+      "length": [
+        "521:5511",
+        0
+      ],
+      "batch_size": 1
+    },
+    "class_type": "EmptyLTXVLatentVideo",
+    "_meta": {
+      "title": "EmptyLTXVLatentVideo"
+    }
+  },
+  "521:473": {
+    "inputs": {
+      "image": [
+        "521:472",
+        0
+      ]
+    },
+    "class_type": "GetImageSize",
+    "_meta": {
+      "title": "Get Image Size"
+    }
+  },
+  "521:474": {
+    "inputs": {
+      "video_latent": [
+        "521:470",
+        0
+      ],
+      "audio_latent": [
+        "521:503",
+        0
+      ]
+    },
+    "class_type": "LTXVConcatAVLatent",
+    "_meta": {
+      "title": "LTXVConcatAVLatent"
+    }
+  },
+  "521:475": {
+    "inputs": {
+      "sigmas": "0.8025, 0.6332, 0.3425, 0.0"
+    },
+    "class_type": "ManualSigmas",
+    "_meta": {
+      "title": "ManualSigmas"
+    }
+  },
+  "521:476": {
+    "inputs": {
+      "video_latent": [
+        "521:495",
+        0
+      ],
+      "audio_latent": [
+        "521:519",
+        1
+      ]
+    },
+    "class_type": "LTXVConcatAVLatent",
+    "_meta": {
+      "title": "LTXVConcatAVLatent"
+    }
+  },
+  "521:478": {
+    "inputs": {
+      "noise": [
+        "521:5542",
+        0
+      ],
+      "guider": [
+        "521:498",
+        0
+      ],
+      "sampler": [
+        "521:464",
+        0
+      ],
+      "sigmas": [
+        "521:465",
+        0
+      ],
+      "latent_image": [
+        "521:474",
+        0
+      ]
+    },
+    "class_type": "SamplerCustomAdvanced",
+    "_meta": {
+      "title": "SamplerCustomAdvanced"
+    }
+  },
+  "521:486": {
+    "inputs": {
+      "longer_edge": 1024,
+      "images": [
+        "5429",
+        0
+      ]
+    },
+    "class_type": "ResizeImagesByLongerEdge",
+    "_meta": {
+      "title": "Resize Images by Longer Edge"
+    }
+  },
+  "521:495": {
+    "inputs": {
+      "strength": 0.8,
+      "bypass": false,
+      "vae": [
+        "174",
+        0
+      ],
+      "image": [
+        "521:492",
+        0
+      ],
+      "latent": [
+        "521:477",
+        0
+      ]
+    },
+    "class_type": "LTXVImgToVideoInplace",
+    "_meta": {
+      "title": "LTXVImgToVideoInplace"
+    }
+  },
+  "521:503": {
+    "inputs": {
+      "samples": [
+        "521:510",
+        0
+      ],
+      "mask": [
+        "521:469",
+        0
+      ]
+    },
+    "class_type": "SetLatentNoiseMask",
+    "_meta": {
+      "title": "Set Latent Noise Mask"
+    }
+  },
+  "521:517": {
+    "inputs": {
+      "frame_rate": [
+        "521:5513",
+        0
+      ],
+      "positive": [
+        "5536",
+        0
+      ],
+      "negative": [
+        "5537",
+        0
+      ]
+    },
+    "class_type": "LTXVConditioning",
+    "_meta": {
+      "title": "LTXVConditioning"
+    }
+  },
+  "521:518": {
+    "inputs": {
+      "positive": [
+        "521:517",
+        0
+      ],
+      "negative": [
+        "521:517",
+        1
+      ],
+      "latent": [
+        "521:519",
+        0
+      ]
+    },
+    "class_type": "LTXVCropGuides",
+    "_meta": {
+      "title": "LTXVCropGuides"
+    }
+  },
+  "521:519": {
+    "inputs": {
+      "av_latent": [
+        "521:478",
+        0
+      ]
+    },
+    "class_type": "LTXVSeparateAVLatent",
+    "_meta": {
+      "title": "LTXVSeparateAVLatent"
+    }
+  },
+  "521:464": {
+    "inputs": {
+      "sampler_name": "euler_ancestral"
+    },
+    "class_type": "KSamplerSelect",
+    "_meta": {
+      "title": "KSamplerSelect"
+    }
+  },
+  "521:466": {
+    "inputs": {
+      "cfg": 1,
+      "model": [
+        "521:606",
+        0
+      ],
+      "positive": [
+        "521:518",
+        0
+      ],
+      "negative": [
+        "521:518",
+        1
+      ]
+    },
+    "class_type": "CFGGuider",
+    "_meta": {
+      "title": "CFGGuider"
+    }
+  },
+  "521:498": {
+    "inputs": {
+      "cfg": 1,
+      "model": [
+        "521:606",
+        0
+      ],
+      "positive": [
+        "521:517",
+        0
+      ],
+      "negative": [
+        "521:517",
+        1
+      ]
+    },
+    "class_type": "CFGGuider",
+    "_meta": {
+      "title": "CFGGuider"
+    }
+  },
+  "521:470": {
+    "inputs": {
+      "strength": 0.8,
+      "bypass": false,
+      "vae": [
+        "174",
+        0
+      ],
+      "image": [
+        "521:492",
+        0
+      ],
+      "latent": [
+        "521:471",
+        0
+      ]
+    },
+    "class_type": "LTXVImgToVideoInplace",
+    "_meta": {
+      "title": "LTXVImgToVideoInplace"
+    }
+  },
+  "521:468": {
+    "inputs": {
+      "noise": [
+        "521:5542",
+        0
+      ],
+      "guider": [
+        "521:466",
+        0
+      ],
+      "sampler": [
+        "521:464",
+        0
+      ],
+      "sigmas": [
+        "521:475",
+        0
+      ],
+      "latent_image": [
+        "521:476",
+        0
+      ]
+    },
+    "class_type": "SamplerCustomAdvanced",
+    "_meta": {
+      "title": "SamplerCustomAdvanced"
+    }
+  },
+  "521:477": {
+    "inputs": {
+      "samples": [
+        "521:519",
+        0
+      ],
+      "upscale_model": [
+        "101",
+        0
+      ],
+      "vae": [
+        "174",
+        0
+      ]
+    },
+    "class_type": "LTXVLatentUpsampler",
+    "_meta": {
+      "title": "LTXVLatentUpsampler"
+    }
+  },
+  "521:522": {
+    "inputs": {
+      "tile_size": 512,
+      "overlap": 64,
+      "temporal_size": 2048,
+      "temporal_overlap": 8,
+      "samples": [
+        "521:479",
+        0
+      ],
+      "vae": [
+        "174",
+        0
+      ]
+    },
+    "class_type": "VAEDecodeTiled",
+    "_meta": {
+      "title": "VAE Decode (Tiled)"
+    }
+  },
+  "521:606": {
+    "inputs": {
+      "preview_rate": 8,
+      "model": [
+        "700",
+        0
+      ],
+      "vae": [
+        "591",
+        0
+      ]
+    },
+    "class_type": "LTX2SamplingPreviewOverride",
+    "_meta": {
+      "title": "LTX2 Sampling Preview Override"
+    }
+  },
+  "521:492": {
+    "inputs": {
+      "img_compression": 33,
+      "image": [
+        "521:486",
+        0
+      ]
+    },
+    "class_type": "LTXVPreprocess",
+    "_meta": {
+      "title": "LTXVPreprocess"
+    }
+  },
+  "521:513": {
+    "inputs": {
+      "images": [
+        "521:522",
+        0
+      ]
+    },
+    "class_type": "FinalFrameSelector",
+    "_meta": {
+      "title": "Final Frame Selector"
+    }
+  },
+  "521:485": {
+    "inputs": {
+      "width": [
+        "5383",
+        0
+      ],
+      "height": [
+        "5382",
+        0
+      ],
+      "batch_size": 1,
+      "color": 0
+    },
+    "class_type": "EmptyImage",
+    "_meta": {
+      "title": "EmptyImage"
+    }
+  },
+  "521:5513": {
+    "inputs": {
+      "value": [
+        "5446",
+        0
+      ]
+    },
+    "class_type": "PrimitiveFloat",
+    "_meta": {
+      "title": "fps"
+    }
+  },
+  "521:479": {
+    "inputs": {
+      "av_latent": [
+        "521:468",
+        0
+      ]
+    },
+    "class_type": "LTXVSeparateAVLatent",
+    "_meta": {
+      "title": "LTXVSeparateAVLatent"
+    }
+  },
+  "521:472": {
+    "inputs": {
+      "upscale_method": "lanczos",
+      "scale_by": 0.5,
+      "image": [
+        "521:485",
+        0
+      ]
+    },
+    "class_type": "ImageScaleBy",
+    "_meta": {
+      "title": "Upscale Image By"
+    }
+  },
+  "521:510": {
+    "inputs": {
+      "audio": [
+        "5400",
+        0
+      ],
+      "audio_vae": [
+        "175",
+        0
+      ]
+    },
+    "class_type": "LTXVAudioVAEEncode",
+    "_meta": {
+      "title": "LTXV Audio VAE Encode"
+    }
+  },
+  "521:5511": {
+    "inputs": {
+      "expression": "a*b+1",
+      "a": [
+        "196",
+        0
+      ],
+      "b": [
+        "521:5513",
+        0
+      ]
+    },
+    "class_type": "MathExpression|pysssss",
+    "_meta": {
+      "title": "Math Expression 🐍"
+    }
+  },
+  "521:5512": {
+    "inputs": {
+      "a": [
+        "521:5513",
+        0
+      ]
+    },
+    "class_type": "CM_FloatToInt",
+    "_meta": {
+      "title": "FloatToInt"
+    }
+  },
+  "521:5542": {
+    "inputs": {
+      "noise_seed": 87299332486566
+    },
+    "class_type": "RandomNoise",
+    "_meta": {
+      "title": "RandomNoise"
+    }
+  }
+}