Skip to content

LTX-Video Support for ComfyUI

Automated PR - 2025-04-17

sync-botbdde649e2506
diff --git a/README.md b/README.md
index eebc262..81c7717 100644
--- a/README.md
+++ b/README.md
@@ -4,7 +4,19 @@ ComfyUI-LTXVideo is a collection of custom nodes for ComfyUI, designed to provid
The model itself is supported in the core ComfyUI [code](https://github.com/comfyanonymous/ComfyUI/tree/master/comfy/ldm/lightricks).
The main LTXVideo repository can be found [here](https://github.com/Lightricks/LTX-Video).
-# 5.03.2025 :star: LTXVideo 0.9.5 Release :star:
+# 17.04.2025 ⭐ LTXVideo 0.9.6 Release ⭐
+
+### LTXVideo 0.9.6 introduces:
+
+1. LTXV 0.9.6 – higher quality, faster, great for final output. Download from [here](https://huggingface.co/Lightricks/LTX-Video/resolve/main/ltxv-2b-0.9.6-dev-04-25.safetensors).
+2. LTXV 0.9.6 Distilled – our fastest model yet (only 8 steps for generation), lighter, great for rapid iteration. Download from [here](https://huggingface.co/Lightricks/LTX-Video/resolve/main/ltxv-2b-0.9.6-distilled-04-25.safetensors).
+
+### Technical Updates
+
+We introduce the __STGGuiderAdvanced__ node, which applies different CFG and STG parameters at various diffusion steps. All flows have been updated to use this node and are designed to provide optimal parameters for the best quality.
+See the [Example Workflows](#example-workflows) section.
+
+# 5.03.2025 ⭐ LTXVideo 0.9.5 Release ⭐
### LTXVideo 0.9.5 introduces:
@@ -73,6 +85,10 @@ Note that to run the example workflows, you need to have some additional custom
[Download workflow](assets/ltxvideo-i2v.json)
![workflow](assets/ltxvideo-i2v.png)
+### Image-to-video-distilled
+[Download workflow](assets/ltxvideo-i2v-distilled.json)
+![workflow](assets/ltxvideo-i2v-distilled.png)
+
### Text-to-video
[Download workflow](assets/ltxvideo-t2v.json)
diff --git a/__init__.py b/__init__.py
index 5bb1b44..42aa9e3 100644
--- a/__init__.py
+++ b/__init__.py
@@ -1,9 +1,14 @@
+from .latents import LTXVSelectLatents, LTXVSetVideoLatentNoiseMasks # noqa: F401
from .nodes_registry import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS
from .prompt_enhancer_nodes import ( # noqa: F401
LTXVPromptEnhancer,
LTXVPromptEnhancerLoader,
)
-from .stg import LTXVApplySTG, STGGuiderNode # noqa: F401
+from .recurrent_sampler import ( # noqa: F401
+ LinearOverlapLatentTransition,
+ LTXVRecurrentKSampler,
+)
+from .stg import LTXVApplySTG, STGGuiderAdvancedNode, STGGuiderNode # noqa: F401
from .tricks import NODE_CLASS_MAPPINGS as TRICKS_NODE_CLASS_MAPPINGS
from .tricks import NODE_DISPLAY_NAME_MAPPINGS as TRICKS_NODE_DISPLAY_NAME_MAPPINGS
diff --git a/assets/fox.jpg b/assets/fox.jpg
new file mode 100644
index 0000000..f766a9f
Binary files /dev/null and b/assets/fox.jpg differ
diff --git a/assets/ltxvideo-first-sequence-conditioning.json b/assets/ltxvideo-first-sequence-conditioning.json
index 0069920..6886662 100644
--- a/assets/ltxvideo-first-sequence-conditioning.json
+++ b/assets/ltxvideo-first-sequence-conditioning.json
@@ -1,85 +1,7 @@
{
- "last_node_id": 212,
- "last_link_id": 579,
+ "last_node_id": 213,
+ "last_link_id": 583,
"nodes": [
- {
- "id": 195,
- "type": "VHS_VideoCombine",
- "pos": [
- 1909.384765625,
- 19.35366439819336
- ],
- "size": [
- 315,
- 544.6666870117188
- ],
- "flags": {},
- "order": 25,
- "mode": 0,
- "inputs": [
- {
- "name": "images",
- "type": "IMAGE",
- "shape": 7,
- "link": 530
- },
- {
- "name": "audio",
- "type": "AUDIO",
- "shape": 7,
- "link": null
- },
- {
- "name": "meta_batch",
- "type": "VHS_BatchManager",
- "shape": 7,
- "link": null
- },
- {
- "name": "vae",
- "type": "VAE",
- "shape": 7,
- "link": null
- }
- ],
- "outputs": [
- {
- "name": "Filenames",
- "type": "VHS_FILENAMES",
- "links": null
- }
- ],
- "properties": {
- "cnr_id": "comfyui-videohelpersuite",
- "ver": "8629188458dc6cb832f871ece3bd273507e8a766",
- "Node name for S&R": "VHS_VideoCombine"
- },
- "widgets_values": {
- "frame_rate": 25,
- "loop_count": 0,
- "filename_prefix": "ltxv",
- "format": "video/h264-mp4",
- "pix_fmt": "yuv420p",
- "crf": 19,
- "save_metadata": true,
- "trim_to_audio": false,
- "pingpong": false,
- "save_output": true,
- "videopreview": {
- "hidden": false,
- "paused": false,
- "params": {
- "filename": "ltxv_00032.mp4",
- "subfolder": "",
- "type": "output",
- "format": "video/h264-mp4",
- "frame_rate": 25,
- "workflow": "ltxv_00032.png",
- "fullpath": "/home/akvochko/ComfyUI-Orig/output/ltxv_00032.mp4"
- }
- }
- }
- },
{
"id": 71,
"type": "LTXVScheduler",
@@ -92,11 +14,12 @@
154
],
"flags": {},
- "order": 20,
+ "order": 19,
"mode": 0,
"inputs": [
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"shape": 7,
"link": 564
@@ -105,6 +28,7 @@
"outputs": [
{
"name": "SIGMAS",
+ "localized_name": "SIGMAS",
"type": "SIGMAS",
"links": [
542
@@ -113,9 +37,9 @@
}
],
"properties": {
+ "Node name for S&R": "LTXVScheduler",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "LTXVScheduler"
+ "ver": "0.3.19"
},
"widgets_values": [
20,
@@ -125,80 +49,6 @@
0.1
]
},
- {
- "id": 188,
- "type": "LTXVApplySTG",
- "pos": [
- 77.9090805053711,
- -126.5878677368164
- ],
- "size": [
- 315,
- 58
- ],
- "flags": {},
- "order": 8,
- "mode": 0,
- "inputs": [
- {
- "name": "model",
- "type": "MODEL",
- "link": 508
- }
- ],
- "outputs": [
- {
- "name": "model",
- "type": "MODEL",
- "links": [
- 516
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "aux_id": "Lightricks/ComfyUI-LTXVideo",
- "ver": "55f2d9ebb885a1bda63f1757080b4d7f8f7bb2f9",
- "Node name for S&R": "LTXVApplySTG"
- },
- "widgets_values": [
- "14"
- ]
- },
- {
- "id": 73,
- "type": "KSamplerSelect",
- "pos": [
- 471.1808166503906,
- -136.36605834960938
- ],
- "size": [
- 315,
- 58
- ],
- "flags": {},
- "order": 0,
- "mode": 0,
- "inputs": [],
- "outputs": [
- {
- "name": "SAMPLER",
- "type": "SAMPLER",
- "links": [
- 541
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "KSamplerSelect"
- },
- "widgets_values": [
- "gradient_estimation"
- ]
- },
{
"id": 69,
"type": "LTXVConditioning",
@@ -211,16 +61,18 @@
78
],
"flags": {},
- "order": 17,
+ "order": 16,
"mode": 0,
"inputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"link": 266
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"link": 267
}
@@ -228,6 +80,7 @@
"outputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"links": [
556
@@ -236,6 +89,7 @@
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"links": [
557
@@ -244,9 +98,9 @@
}
],
"properties": {
+ "Node name for S&R": "LTXVConditioning",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "LTXVConditioning"
+ "ver": "0.3.19"
},
"widgets_values": [
25
@@ -264,12 +118,13 @@
82
],
"flags": {},
- "order": 1,
+ "order": 0,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "NOISE",
+ "localized_name": "NOISE",
"type": "NOISE",
"links": [
540
@@ -278,106 +133,15 @@
}
],
"properties": {
+ "Node name for S&R": "RandomNoise",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "RandomNoise"
+ "ver": "0.3.19"
},
"widgets_values": [
43,
"fixed"
]
},
- {
- "id": 44,
- "type": "CheckpointLoaderSimple",
- "pos": [
- -301.724853515625,
- -148.44154357910156
- ],
- "size": [
- 315,
- 98
- ],
- "flags": {},
- "order": 2,
- "mode": 0,
- "inputs": [],
- "outputs": [
- {
- "name": "MODEL",
- "type": "MODEL",
- "links": [
- 508
- ],
- "slot_index": 0
- },
- {
- "name": "CLIP",
- "type": "CLIP",
- "links": null
- },
- {
- "name": "VAE",
- "type": "VAE",
- "links": [
- 565,
- 566
- ],
- "slot_index": 2
- }
- ],
- "properties": {
- "cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "CheckpointLoaderSimple"
- },
- "widgets_values": [
- "ltx-video-2b-v0.9.5.safetensors"
- ]
- },
- {
- "id": 193,
- "type": "ImageBatch",
- "pos": [
- 1659.6142578125,
- 240.26751708984375
- ],
- "size": [
- 210,
- 46
- ],
- "flags": {},
- "order": 24,
- "mode": 0,
- "inputs": [
- {
- "name": "image1",
- "type": "IMAGE",
- "link": 537
- },
- {
- "name": "image2",
- "type": "IMAGE",
- "link": 536
- }
- ],
- "outputs": [
- {
- "name": "IMAGE",
- "type": "IMAGE",
- "links": [
- 530
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "ImageBatch"
- },
- "widgets_values": []
- },
{
"id": 199,
"type": "LTXVPromptEnhancerLoader",
@@ -390,12 +154,13 @@
82
],
"flags": {},
- "order": 3,
+ "order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "prompt_enhancer",
+ "localized_name": "prompt_enhancer",
"type": "LTXV_PROMPT_ENHANCER",
"links": [
538
@@ -404,9 +169,9 @@
}
],
"properties": {
+ "Node name for S&R": "LTXVPromptEnhancerLoader",
"aux_id": "Lightricks/ComfyUI-LTXVideo",
- "ver": "55f2d9ebb885a1bda63f1757080b4d7f8f7bb2f9",
- "Node name for S&R": "LTXVPromptEnhancerLoader"
+ "ver": "55f2d9ebb885a1bda63f1757080b4d7f8f7bb2f9"
},
"widgets_values": [
"unsloth/Llama-3.2-3B-Instruct",
@@ -425,12 +190,13 @@
98
],
"flags": {},
- "order": 4,
+ "order": 2,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "CLIP",
+ "localized_name": "CLIP",
"type": "CLIP",
"links": [
74,
@@ -440,9 +206,9 @@
}
],
"properties": {
+ "Node name for S&R": "CLIPLoader",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "CLIPLoader"
+ "ver": "0.3.19"
},
"widgets_values": [
"t5xxl_fp16.safetensors",
@@ -464,11 +230,12 @@
"flags": {
"collapsed": true
},
- "order": 14,
+ "order": 13,
"mode": 0,
"inputs": [
{
"name": "clip",
+ "localized_name": "clip",
"type": "CLIP",
"link": 74
},
@@ -484,6 +251,7 @@
"outputs": [
{
"name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
"type": "CONDITIONING",
"links": [
266
@@ -493,9 +261,9 @@
],
"title": "CLIP Text Encode (Positive Prompt)",
"properties": {
+ "Node name for S&R": "CLIPTextEncode",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "CLIPTextEncode"
+ "ver": "0.3.19"
},
"widgets_values": [
"A confident man with short dark hair and a beard, wearing a casual green jacket, sits at a chessboard with a smile. He leans forward, hands near the pieces, as if about to make a move. The background shows a dimly lit hall with a blurred audience watching, creating a tense yet exciting atmosphere. The video should capture subtle movements like eye shifts, hand gestures, and slight changes in facial expression to enhance realism."
@@ -517,11 +285,12 @@
"flags": {
"collapsed": true
},
- "order": 9,
+ "order": 8,
"mode": 0,
"inputs": [
{
"name": "clip",
+ "localized_name": "clip",
"type": "CLIP",
"link": 75
}
@@ -529,6 +298,7 @@
"outputs": [
{
"name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
"type": "CONDITIONING",
"links": [
267
@@ -538,9 +308,9 @@
],
"title": "CLIP Text Encode (Negative Prompt)",
"properties": {
+ "Node name for S&R": "CLIPTextEncode",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "CLIPTextEncode"
+ "ver": "0.3.19"
},
"widgets_values": [
"low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly"
@@ -560,16 +330,18 @@
102
],
"flags": {},
- "order": 10,
+ "order": 9,
"mode": 0,
"inputs": [
{
"name": "prompt_enhancer",
+ "localized_name": "prompt_enhancer",
"type": "LTXV_PROMPT_ENHANCER",
"link": 538
},
{
"name": "image_prompt",
+ "localized_name": "image_prompt",
"type": "IMAGE",
"shape": 7,
"link": 523
@@ -578,6 +350,7 @@
"outputs": [
{
"name": "str",
+ "localized_name": "str",
"type": "STRING",
"links": [
507
@@ -586,9 +359,9 @@
}
],
"properties": {
+ "Node name for S&R": "LTXVPromptEnhancer",
"aux_id": "Lightricks/ComfyUI-LTXVideo",
- "ver": "55f2d9ebb885a1bda63f1757080b4d7f8f7bb2f9",
- "Node name for S&R": "LTXVPromptEnhancer"
+ "ver": "55f2d9ebb885a1bda63f1757080b4d7f8f7bb2f9"
},
"widgets_values": [
"red jeep",
@@ -607,11 +380,12 @@
86
],
"flags": {},
- "order": 13,
+ "order": 12,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 567
}
@@ -619,11 +393,13 @@
"outputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"links": null
},
{
- "name": "width",
+ "name": "768 width",
+ "localized_name": "width",
"type": "INT",
"links": [
568
@@ -631,7 +407,8 @@
"slot_index": 1
},
{
- "name": "height",
+ "name": "512 height",
+ "localized_name": "height",
"type": "INT",
"links": [
569
@@ -639,7 +416,8 @@
"slot_index": 2
},
{
- "name": "count",
+ "name": "65 count",
+ "localized_name": "count",
"type": "INT",
"links": [
570
@@ -648,9 +426,9 @@
}
],
"properties": {
+ "Node name for S&R": "GetImageSizeAndCount",
"cnr_id": "comfyui-kjnodes",
- "ver": "fa6d20eeb32cbcbd84ee325e3cfdc6ed6b43056f",
- "Node name for S&R": "GetImageSizeAndCount"
+ "ver": "fa6d20eeb32cbcbd84ee325e3cfdc6ed6b43056f"
},
"widgets_values": []
},
@@ -666,11 +444,12 @@
106
],
"flags": {},
- "order": 11,
+ "order": 10,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 525
}
@@ -678,6 +457,7 @@
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
571
@@ -686,9 +466,9 @@
}
],
"properties": {
+ "Node name for S&R": "VHS_SelectImages",
"cnr_id": "comfyui-videohelpersuite",
- "ver": "8629188458dc6cb832f871ece3bd273507e8a766",
- "Node name for S&R": "VHS_SelectImages"
+ "ver": "8629188458dc6cb832f871ece3bd273507e8a766"
},
"widgets_values": {
"indexes": "-9:",
@@ -708,7 +488,7 @@
93.54439544677734
],
"flags": {},
- "order": 5,
+ "order": 3,
"mode": 0,
"inputs": [],
"outputs": [],
@@ -731,7 +511,7 @@
170
],
"flags": {},
- "order": 16,
+ "order": 15,
"mode": 0,
"inputs": [
{
@@ -774,6 +554,7 @@
"outputs": [
{
"name": "LATENT",
+ "localized_name": "LATENT",
"type": "LATENT",
"links": [
558
@@ -782,9 +563,9 @@
}
],
"properties": {
+ "Node name for S&R": "EmptyLTXVLatentVideo",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "EmptyLTXVLatentVideo"
+ "ver": "0.3.19"
},
"widgets_values": [
768,
@@ -805,11 +586,12 @@
58
],
"flags": {},
- "order": 15,
+ "order": 14,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 571
}
@@ -817,6 +599,7 @@
"outputs": [
{
"name": "output_image",
+ "localized_name": "output_image",
"type": "IMAGE",
"links": [
572
@@ -825,9 +608,9 @@
}
],
"properties": {
+ "Node name for S&R": "LTXVPreprocess",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "LTXVPreprocess"
+ "ver": "0.3.19"
},
"widgets_values": [
35
@@ -845,11 +628,12 @@
106
],
"flags": {},
- "order": 12,
+ "order": 11,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 527
}
@@ -857,6 +641,7 @@
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
537
@@ -865,9 +650,9 @@
}
],
"properties": {
+ "Node name for S&R": "VHS_SelectImages",
"cnr_id": "comfyui-videohelpersuite",
- "ver": "8629188458dc6cb832f871ece3bd273507e8a766",
- "Node name for S&R": "VHS_SelectImages"
+ "ver": "8629188458dc6cb832f871ece3bd273507e8a766"
},
"widgets_values": {
"indexes": ":-9",
@@ -887,7 +672,7 @@
93.54439544677734
],
"flags": {},
- "order": 6,
+ "order": 4,
"mode": 0,
"inputs": [],
"outputs": [],
@@ -907,20 +692,22 @@
],
"size": [
252.056640625,
- 478.10009765625
+ 430.7044372558594
],
"flags": {},
- "order": 7,
+ "order": 5,
"mode": 0,
"inputs": [
{
"name": "meta_batch",
+ "localized_name": "meta_batch",
"type": "VHS_BatchManager",
"shape": 7,
"link": null
},
{
"name": "vae",
+ "localized_name": "vae",
"type": "VAE",
"shape": 7,
"link": null
@@ -929,6 +716,7 @@
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
523,
@@ -940,34 +728,37 @@
},
{
"name": "frame_count",
+ "localized_name": "frame_count",
"type": "INT",
"links": null
},
{
"name": "audio",
+ "localized_name": "audio",
"type": "AUDIO",
"links": null
},
{
"name": "video_info",
+ "localized_name": "video_info",
"type": "VHS_VIDEOINFO",
"links": null
}
],
"properties": {
+ "Node name for S&R": "VHS_LoadVideo",
"cnr_id": "comfyui-videohelpersuite",
- "ver": "8629188458dc6cb832f871ece3bd273507e8a766",
- "Node name for S&R": "VHS_LoadVideo"
+ "ver": "8629188458dc6cb832f871ece3bd273507e8a766"
},
"widgets_values": {
"video": "jeep.mp4",
"force_rate": 0,
+ "force_size": "Disabled",
"custom_width": 0,
"custom_height": 0,
"frame_load_cap": 65,
"skip_first_frames": 0,
"select_every_nth": 1,
- "format": "ltxv",
"choose video to upload": "image",
"videopreview": {
"hidden": false,
@@ -985,98 +776,54 @@
}
},
{
- "id": 8,
- "type": "VAEDecode",
+ "id": 44,
+ "type": "CheckpointLoaderSimple",
"pos": [
- 1658.79638671875,
- 121.616455078125
+ -301.724853515625,
+ -148.44154357910156
],
"size": [
- 210,
- 46
+ 315,
+ 98
],
"flags": {},
- "order": 23,
+ "order": 6,
"mode": 0,
- "inputs": [
- {
- "name": "samples",
- "type": "LATENT",
- "link": 579
- },
- {
- "name": "vae",
- "type": "VAE",
- "link": 565
- }
- ],
+ "inputs": [],
"outputs": [
{
- "name": "IMAGE",
- "type": "IMAGE",
+ "name": "MODEL",
+ "localized_name": "MODEL",
+ "type": "MODEL",
"links": [
- 536
+ 580
],
"slot_index": 0
- }
- ],
- "properties": {
- "cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "VAEDecode"
- },
- "widgets_values": []
- },
- {
- "id": 191,
- "type": "STGGuider",
- "pos": [
- 1216.2901611328125,
- -1.7041345834732056
- ],
- "size": [
- 389.04827880859375,
- 171.58651733398438
- ],
- "flags": {},
- "order": 19,
- "mode": 0,
- "inputs": [
- {
- "name": "model",
- "type": "MODEL",
- "link": 516
},
{
- "name": "positive",
- "type": "CONDITIONING",
- "link": 576
+ "name": "CLIP",
+ "localized_name": "CLIP",
+ "type": "CLIP",
+ "links": null
},
{
- "name": "negative",
- "type": "CONDITIONING",
- "link": 577
- }
- ],
- "outputs": [
- {
- "name": "GUIDER",
- "type": "GUIDER",
+ "name": "VAE",
+ "localized_name": "VAE",
+ "type": "VAE",
"links": [
- 544
+ 565,
+ 566
],
- "slot_index": 0
+ "slot_index": 2
}
],
"properties": {
- "aux_id": "Lightricks/ComfyUI-LTXVideo",
- "ver": "55f2d9ebb885a1bda63f1757080b4d7f8f7bb2f9",
- "Node name for S&R": "STGGuider"
+ "Node name for S&R": "CheckpointLoaderSimple",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.19"
},
"widgets_values": [
- 3.5,
- 1,
- 0.75
+ "ltxv-2b-0.9.6-dev-04-25.safetensors"
]
},
{
@@ -1091,31 +838,36 @@
162
],
"flags": {},
- "order": 18,
+ "order": 17,
"mode": 0,
"inputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"link": 556
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"link": 557
},
{
"name": "vae",
+ "localized_name": "vae",
"type": "VAE",
"link": 566
},
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"link": 558
},
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 572
}
@@ -1123,24 +875,27 @@
"outputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"links": [
574,
- 576
+ 581
],
"slot_index": 0
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"links": [
575,
- 577
+ 582
],
"slot_index": 1
},
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"links": [
563,
@@ -1150,9 +905,9 @@
}
],
"properties": {
+ "Node name for S&R": "LTXVAddGuide",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "LTXVAddGuide"
+ "ver": "0.3.19"
},
"widgets_values": [
0,
@@ -1160,95 +915,128 @@
]
},
{
- "id": 200,
- "type": "SamplerCustomAdvanced",
+ "id": 73,
+ "type": "KSamplerSelect",
"pos": [
- 1210.78759765625,
- 282.859130859375
+ 841.9295654296875,
+ 520.01416015625
],
"size": [
- 355.20001220703125,
- 106
+ 315,
+ 58
],
"flags": {},
- "order": 21,
+ "order": 7,
"mode": 0,
- "inputs": [
- {
- "name": "noise",
- "type": "NOISE",
- "link": 540
- },
- {
- "name": "guider",
- "type": "GUIDER",
- "link": 544
- },
+ "inputs": [],
+ "outputs": [
{
- "name": "sampler",
+ "name": "SAMPLER",
+ "localized_name": "SAMPLER",
"type": "SAMPLER",
- "link": 541
+ "links": [
+ 541
+ ],
+ "slot_index": 0
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "KSamplerSelect",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.19"
+ },
+ "widgets_values": [
+ "gradient_estimation"
+ ]
+ },
+ {
+ "id": 213,
+ "type": "STGGuiderAdvanced",
+ "pos": [
+ 1220.742919921875,
+ -132.0933074951172
+ ],
+ "size": [
+ 327.5999755859375,
+ 242
+ ],
+ "flags": {},
+ "order": 18,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "model",
+ "localized_name": "model",
+ "type": "MODEL",
+ "link": 580
},
{
- "name": "sigmas",
- "type": "SIGMAS",
- "link": 542
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 581
},
{
- "name": "latent_image",
- "type": "LATENT",
- "link": 563
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 582
}
],
"outputs": [
{
- "name": "output",
- "type": "LATENT",
+ "name": "GUIDER",
+ "localized_name": "GUIDER",
+ "type": "GUIDER",
"links": [
- 578
+ 583
],
"slot_index": 0
- },
- {
- "name": "denoised_output",
- "type": "LATENT",
- "links": null
}
],
"properties": {
- "cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "SamplerCustomAdvanced"
+ "Node name for S&R": "STGGuiderAdvanced"
},
- "widgets_values": []
+ "widgets_values": [
+ 0.996,
+ true,
+ "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "4, 4, 4, 4, 1, 1",
+ "2, 2, 2, 2, 1, 0",
+ "1, 1, 1, 1, 1, 1",
+ "[14], [14], [14], [14], [14], [14]"
+ ]
},
{
"id": 212,
"type": "LTXVCropGuides",
"pos": [
- 1653.6558837890625,
- 1.2874510288238525
+ 1254.9326171875,
+ 344.5459289550781
],
"size": [
216.59999084472656,
66
],
"flags": {},
- "order": 22,
+ "order": 21,
"mode": 0,
"inputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"link": 574
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"link": 575
},
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"link": 578
}
@@ -1256,16 +1044,19 @@
"outputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"links": null
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"links": null
},
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"links": [
579
@@ -1274,9 +1065,253 @@
}
],
"properties": {
+ "Node name for S&R": "LTXVCropGuides",
"cnr_id": "comfy-core",
- "ver": "0.3.19",
- "Node name for S&R": "LTXVCropGuides"
+ "ver": "0.3.19"
+ },
+ "widgets_values": []
+ },
+ {
+ "id": 8,
+ "type": "VAEDecode",
+ "pos": [
+ 1255.8253173828125,
+ 468.36798095703125
+ ],
+ "size": [
+ 210,
+ 46
+ ],
+ "flags": {},
+ "order": 22,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "samples",
+ "localized_name": "samples",
+ "type": "LATENT",
+ "link": 579
+ },
+ {
+ "name": "vae",
+ "localized_name": "vae",
+ "type": "VAE",
+ "link": 565
+ }
+ ],
+ "outputs": [
+ {
+ "name": "IMAGE",
+ "localized_name": "IMAGE",
+ "type": "IMAGE",
+ "links": [
+ 536
+ ],
+ "slot_index": 0
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "VAEDecode",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.19"
+ },
+ "widgets_values": []
+ },
+ {
+ "id": 193,
+ "type": "ImageBatch",
+ "pos": [
+ 1613.5460205078125,
+ -128.23391723632812
+ ],
+ "size": [
+ 210,
+ 46
+ ],
+ "flags": {},
+ "order": 23,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "image1",
+ "localized_name": "image1",
+ "type": "IMAGE",
+ "link": 537
+ },
+ {
+ "name": "image2",
+ "localized_name": "image2",
+ "type": "IMAGE",
+ "link": 536
+ }
+ ],
+ "outputs": [
+ {
+ "name": "IMAGE",
+ "localized_name": "IMAGE",
+ "type": "IMAGE",
+ "links": [
+ 530
+ ],
+ "slot_index": 0
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "ImageBatch",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.19"
+ },
+ "widgets_values": []
+ },
+ {
+ "id": 195,
+ "type": "VHS_VideoCombine",
+ "pos": [
+ 1602.831298828125,
+ -6.919589996337891
+ ],
+ "size": [
+ 315,
+ 520.6666870117188
+ ],
+ "flags": {},
+ "order": 24,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "images",
+ "localized_name": "images",
+ "type": "IMAGE",
+ "shape": 7,
+ "link": 530
+ },
+ {
+ "name": "audio",
+ "localized_name": "audio",
+ "type": "AUDIO",
+ "shape": 7,
+ "link": null
+ },
+ {
+ "name": "meta_batch",
+ "localized_name": "meta_batch",
+ "type": "VHS_BatchManager",
+ "shape": 7,
+ "link": null
+ },
+ {
+ "name": "vae",
+ "localized_name": "vae",
+ "type": "VAE",
+ "shape": 7,
+ "link": null
+ }
+ ],
+ "outputs": [
+ {
+ "name": "Filenames",
+ "localized_name": "Filenames",
+ "type": "VHS_FILENAMES",
+ "links": null
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "VHS_VideoCombine",
+ "cnr_id": "comfyui-videohelpersuite",
+ "ver": "8629188458dc6cb832f871ece3bd273507e8a766"
+ },
+ "widgets_values": {
+ "frame_rate": 25,
+ "loop_count": 0,
+ "filename_prefix": "ltxv",
+ "format": "video/h264-mp4",
+ "pix_fmt": "yuv420p",
+ "crf": 19,
+ "save_metadata": true,
+ "pingpong": false,
+ "save_output": true,
+ "videopreview": {
+ "hidden": false,
+ "paused": false,
+ "params": {
+ "filename": "ltxv_00120.mp4",
+ "subfolder": "",
+ "type": "output",
+ "format": "video/h264-mp4",
+ "frame_rate": 25,
+ "workflow": "ltxv_00032.png",
+ "fullpath": "/home/akvochko/ComfyUI-Orig/output/ltxv_00032.mp4"
+ }
+ }
+ }
+ },
+ {
+ "id": 200,
+ "type": "SamplerCustomAdvanced",
+ "pos": [
+ 1223.8555908203125,
+ 171.28802490234375
+ ],
+ "size": [
+ 326.99456787109375,
+ 106
+ ],
+ "flags": {},
+ "order": 20,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "noise",
+ "localized_name": "noise",
+ "type": "NOISE",
+ "link": 540
+ },
+ {
+ "name": "guider",
+ "localized_name": "guider",
+ "type": "GUIDER",
+ "link": 583
+ },
+ {
+ "name": "sampler",
+ "localized_name": "sampler",
+ "type": "SAMPLER",
+ "link": 541
+ },
+ {
+ "name": "sigmas",
+ "localized_name": "sigmas",
+ "type": "SIGMAS",
+ "link": 542
+ },
+ {
+ "name": "latent_image",
+ "localized_name": "latent_image",
+ "type": "LATENT",
+ "link": 563
+ }
+ ],
+ "outputs": [
+ {
+ "name": "output",
+ "localized_name": "output",
+ "type": "LATENT",
+ "links": [
+ 578
+ ],
+ "slot_index": 0
+ },
+ {
+ "name": "denoised_output",
+ "localized_name": "denoised_output",
+ "type": "LATENT",
+ "links": null
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "SamplerCustomAdvanced",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.19"
},
"widgets_values": []
}
@@ -1322,22 +1357,6 @@
1,
"STRING"
],
- [
- 508,
- 44,
- 0,
- 188,
- 0,
- "MODEL"
- ],
- [
- 516,
- 188,
- 0,
- 191,
- 0,
- "MODEL"
- ],
[
523,
134,
@@ -1418,14 +1437,6 @@
3,
"SIGMAS"
],
- [
- 544,
- 191,
- 0,
- 200,
- 1,
- "GUIDER"
- ],
[
556,
69,
@@ -1546,22 +1557,6 @@
1,
"CONDITIONING"
],
- [
- 576,
- 206,
- 0,
- 191,
- 1,
- "CONDITIONING"
- ],
- [
- 577,
- 206,
- 1,
- 191,
- 2,
- "CONDITIONING"
- ],
[
578,
200,
@@ -1577,16 +1572,48 @@
8,
0,
"LATENT"
+ ],
+ [
+ 580,
+ 44,
+ 0,
+ 213,
+ 0,
+ "MODEL"
+ ],
+ [
+ 581,
+ 206,
+ 0,
+ 213,
+ 1,
+ "CONDITIONING"
+ ],
+ [
+ 582,
+ 206,
+ 1,
+ 213,
+ 2,
+ "CONDITIONING"
+ ],
+ [
+ 583,
+ 213,
+ 0,
+ 200,
+ 1,
+ "GUIDER"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
- "scale": 0.6727499949325706,
+ "scale": 0.6303940863128502,
"offset": [
- 841.6492448928334,
- 316.8380884724249
+ 614.0833708673383,
+ 371.5923298111381
]
},
"prompt": {
@@ -1622,8 +1649,8 @@
"8": {
"inputs": {
"samples": [
- "200",
- 0
+ "212",
+ 2
],
"vae": [
"44",
@@ -1648,7 +1675,7 @@
},
"44": {
"inputs": {
- "ckpt_name": "ltx-video-2b-v0.9.5.safetensors"
+ "ckpt_name": "ltxv-2b-0.9.6-dev-04-25.safetensors"
},
"class_type": "CheckpointLoaderSimple",
"_meta": {
@@ -1724,8 +1751,8 @@
"video": "jeep.mp4",
"force_rate": 0,
"force_size": "Disabled",
- "custom_width": 512,
- "custom_height": 512,
+ "custom_width": 0,
+ "custom_height": 0,
"frame_load_cap": 65,
"skip_first_frames": 0,
"select_every_nth": 1
@@ -1753,19 +1780,6 @@
"title": "🅛🅣🅧 LTXV Prompt Enhancer"
}
},
- "188": {
- "inputs": {
- "block_indices": "14",
- "model": [
- "44",
- 0
- ]
- },
- "class_type": "LTXVApplySTG",
- "_meta": {
- "title": "🅛🅣🅧 LTXV Apply STG"
- }
- },
"190": {
"inputs": {
"noise_seed": 43
@@ -1775,29 +1789,6 @@
"title": "RandomNoise"
}
},
- "191": {
- "inputs": {
- "cfg": 3.5,
- "stg": 1,
- "rescale": 0.75,
- "model": [
- "188",
- 0
- ],
- "positive": [
- "206",
- 0
- ],
- "negative": [
- "206",
- 1
- ]
- },
- "class_type": "STGGuider",
- "_meta": {
- "title": "🅛🅣🅧 STG Guider"
- }
- },
"192": {
"inputs": {
"indexes": "-9:",
@@ -1882,7 +1873,7 @@
0
],
"guider": [
- "191",
+ "213",
0
],
"sampler": [
@@ -1957,11 +1948,58 @@
"_meta": {
"title": "LTXVPreprocess"
}
+ },
+ "212": {
+ "inputs": {
+ "positive": [
+ "206",
+ 0
+ ],
+ "negative": [
+ "206",
+ 1
+ ],
+ "latent": [
+ "200",
+ 0
+ ]
+ },
+ "class_type": "LTXVCropGuides",
+ "_meta": {
+ "title": "LTXVCropGuides"
+ }
+ },
+ "213": {
+ "inputs": {
+ "skip_steps_sigma_threshold": 0.996,
+ "cfg_star_rescale": true,
+ "sigmas": "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "cfg_values": "4, 4, 4, 4, 1, 1",
+ "stg_scale_values": "2, 2, 2, 2, 1, 0",
+ "stg_rescale_values": "1, 1, 1, 1, 1, 1",
+ "stg_layers_indices": "[14], [14], [14], [14], [14], [14]",
+ "model": [
+ "44",
+ 0
+ ],
+ "positive": [
+ "206",
+ 0
+ ],
+ "negative": [
+ "206",
+ 1
+ ]
+ },
+ "class_type": "STGGuiderAdvanced",
+ "_meta": {
+ "title": "🅛🅣🅧 STG Guider Advanced"
+ }
}
},
- "comfy_fork_version": "HEAD@e7bd1751",
+ "comfy_fork_version": "feature/memory_prof@f7e64946",
"node_versions": {
- "comfy-core": "0.3.18"
+ "comfy-core": "0.3.27"
},
"VHS_latentpreview": false,
"VHS_latentpreviewrate": 0,
diff --git a/assets/ltxvideo-first-sequence-conditioning.png b/assets/ltxvideo-first-sequence-conditioning.png
index 0f7a6bf..c43a999 100644
--- a/assets/ltxvideo-first-sequence-conditioning.png
+++ b/assets/ltxvideo-first-sequence-conditioning.png
@@ -1,3 +1,3 @@
version https://git-lfs.github.com/spec/v1
-oid sha256:adab606ec23e98c5aadb7adbfa9d16cb1e9d460004bbe648cb03f5e89972229a
-size 1267760
+oid sha256:7899cb4e0e23f2439cc489b1ed6191eef91e91dbefdcb2c4c29545d5a8cd6c3a
+size 1292625
diff --git a/assets/ltxvideo-frame-interpolation.json b/assets/ltxvideo-frame-interpolation.json
index c86d77c..6144696 100644
--- a/assets/ltxvideo-frame-interpolation.json
+++ b/assets/ltxvideo-frame-interpolation.json
@@ -1,6 +1,6 @@
{
- "last_node_id": 37,
- "last_link_id": 88,
+ "last_node_id": 38,
+ "last_link_id": 93,
"nodes": [
{
"id": 26,
@@ -14,31 +14,36 @@
162
],
"flags": {},
- "order": 17,
+ "order": 15,
"mode": 0,
"inputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"link": 59
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"link": 60
},
{
"name": "vae",
+ "localized_name": "vae",
"type": "VAE",
"link": 74
},
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"link": 62
},
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 63
}
@@ -46,6 +51,7 @@
"outputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"links": [
66
@@ -54,6 +60,7 @@
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"links": [
67
@@ -62,6 +69,7 @@
},
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"links": [
68
@@ -89,11 +97,12 @@
58
],
"flags": {},
- "order": 15,
+ "order": 13,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 81
}
@@ -101,6 +110,7 @@
"outputs": [
{
"name": "output_image",
+ "localized_name": "output_image",
"type": "IMAGE",
"links": [
63
@@ -127,11 +137,12 @@
58
],
"flags": {},
- "order": 16,
+ "order": 14,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 83
}
@@ -139,6 +150,7 @@
"outputs": [
{
"name": "output_image",
+ "localized_name": "output_image",
"type": "IMAGE",
"links": [
69
@@ -154,53 +166,60 @@
]
},
{
- "id": 11,
- "type": "LTXVApplySTG",
+ "id": 31,
+ "type": "LTXVScheduler",
"pos": [
- 1500.3453369140625,
- 1079.3546142578125
+ 1889.956787109375,
+ 1356.0611572265625
],
"size": [
315,
- 58
+ 154
],
"flags": {},
- "order": 12,
+ "order": 18,
"mode": 0,
"inputs": [
{
- "name": "model",
- "type": "MODEL",
- "link": 9
+ "name": "latent",
+ "localized_name": "latent",
+ "type": "LATENT",
+ "shape": 7,
+ "link": 78
}
],
"outputs": [
{
- "name": "model",
- "type": "MODEL",
+ "name": "SIGMAS",
+ "localized_name": "SIGMAS",
+ "type": "SIGMAS",
"links": [
- 10
+ 79
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "LTXVApplySTG"
+ "Node name for S&R": "LTXVScheduler"
},
"widgets_values": [
- "14"
+ 20,
+ 2.05,
+ 0.95,
+ true,
+ 0.1
]
},
{
- "id": 9,
- "type": "KSamplerSelect",
+ "id": 2,
+ "type": "LoadImage",
"pos": [
- 2233.98828125,
- 1065.9569091796875
+ 1091.1929931640625,
+ 1840.934814453125
],
"size": [
315,
- 58
+ 314
],
"flags": {},
"order": 0,
@@ -208,644 +227,682 @@
"inputs": [],
"outputs": [
{
- "name": "SAMPLER",
- "type": "SAMPLER",
+ "name": "IMAGE",
+ "localized_name": "IMAGE",
+ "type": "IMAGE",
"links": [
- 4
+ 80
],
"slot_index": 0
+ },
+ {
+ "name": "MASK",
+ "localized_name": "MASK",
+ "type": "MASK",
+ "links": null
}
],
"properties": {
- "Node name for S&R": "KSamplerSelect"
+ "Node name for S&R": "LoadImage"
},
"widgets_values": [
- "gradient_estimation"
+ "start.jpg",
+ "image"
]
},
{
- "id": 7,
- "type": "SamplerCustomAdvanced",
+ "id": 1,
+ "type": "LoadImage",
"pos": [
- 2620.427978515625,
- 1164.3114013671875
+ 1502.8370361328125,
+ 1844.201171875
],
"size": [
- 355.20001220703125,
- 106
+ 315,
+ 314
],
"flags": {},
- "order": 22,
+ "order": 1,
"mode": 0,
- "inputs": [
- {
- "name": "noise",
- "type": "NOISE",
- "link": 3
- },
- {
- "name": "guider",
- "type": "GUIDER",
- "link": 2
- },
- {
- "name": "sampler",
- "type": "SAMPLER",
- "link": 4
- },
- {
- "name": "sigmas",
- "type": "SIGMAS",
- "link": 79
- },
- {
- "name": "latent_image",
- "type": "LATENT",
- "link": 72
- }
- ],
+ "inputs": [],
"outputs": [
{
- "name": "output",
- "type": "LATENT",
+ "name": "IMAGE",
+ "localized_name": "IMAGE",
+ "type": "IMAGE",
"links": [
- 85
+ 82
],
"slot_index": 0
},
{
- "name": "denoised_output",
- "type": "LATENT",
+ "name": "MASK",
+ "localized_name": "MASK",
+ "type": "MASK",
"links": null
}
],
"properties": {
- "Node name for S&R": "SamplerCustomAdvanced"
+ "Node name for S&R": "LoadImage"
},
- "widgets_values": []
+ "widgets_values": [
+ "end.jpg",
+ "image"
+ ]
},
{
- "id": 5,
- "type": "STGGuider",
+ "id": 35,
+ "type": "ImageResizeKJ",
"pos": [
- 2242.097900390625,
- 1185.6693115234375
+ 1499.1163330078125,
+ 1531.5040283203125
],
"size": [
315,
- 146
+ 286
],
"flags": {},
- "order": 21,
+ "order": 10,
"mode": 0,
"inputs": [
{
- "name": "model",
- "type": "MODEL",
- "link": 10
+ "name": "image",
+ "localized_name": "image",
+ "type": "IMAGE",
+ "link": 82
},
{
- "name": "positive",
- "type": "CONDITIONING",
- "link": 7
+ "name": "get_image_size",
+ "localized_name": "get_image_size",
+ "type": "IMAGE",
+ "shape": 7,
+ "link": null
},
{
- "name": "negative",
- "type": "CONDITIONING",
- "link": 8
+ "name": "width_input",
+ "type": "INT",
+ "shape": 7,
+ "pos": [
+ 10,
+ 196
+ ],
+ "widget": {
+ "name": "width_input"
+ },
+ "link": null
+ },
+ {
+ "name": "height_input",
+ "type": "INT",
+ "shape": 7,
+ "pos": [
+ 10,
+ 220
+ ],
+ "widget": {
+ "name": "height_input"
+ },
+ "link": null
}
],
"outputs": [
{
- "name": "GUIDER",
- "type": "GUIDER",
+ "name": "IMAGE",
+ "localized_name": "IMAGE",
+ "type": "IMAGE",
"links": [
- 2
+ 83
],
"slot_index": 0
+ },
+ {
+ "name": "width",
+ "localized_name": "width",
+ "type": "INT",
+ "links": null
+ },
+ {
+ "name": "height",
+ "localized_name": "height",
+ "type": "INT",
+ "links": null
}
],
"properties": {
- "Node name for S&R": "STGGuider"
+ "Node name for S&R": "ImageResizeKJ"
},
"widgets_values": [
- 3,
- 1,
- 0.75
+ 768,
+ 512,
+ "nearest-exact",
+ false,
+ 2,
+ 0,
+ 0,
+ "center"
]
},
{
- "id": 13,
- "type": "EmptyLTXVLatentVideo",
+ "id": 28,
+ "type": "Note",
"pos": [
- 676.87158203125,
- 1749.301513671875
+ 1095.3414306640625,
+ 2204.1455078125
],
"size": [
- 315,
- 130
+ 314.0157470703125,
+ 91.49161529541016
],
"flags": {},
- "order": 1,
+ "order": 2,
"mode": 0,
"inputs": [],
- "outputs": [
- {
- "name": "LATENT",
- "type": "LATENT",
- "links": [
- 62
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "Node name for S&R": "EmptyLTXVLatentVideo"
- },
+ "outputs": [],
+ "properties": {},
"widgets_values": [
- 768,
- 512,
- 121,
- 1
- ]
+ "First frame conditioning, the frame size shoudl match the generated video aspect ration."
+ ],
+ "color": "#432",
+ "bgcolor": "#653"
},
{
- "id": 15,
- "type": "CLIPTextEncode",
+ "id": 29,
+ "type": "Note",
"pos": [
- 622.5274047851562,
- 1480.2808837890625
+ 1510.7900390625,
+ 2206.3818359375
],
"size": [
- 392.1722717285156,
- 119.6710433959961
+ 304.03729248046875,
+ 88
],
"flags": {},
- "order": 11,
+ "order": 3,
"mode": 0,
- "inputs": [
- {
- "name": "clip",
- "type": "CLIP",
- "link": 19
- }
- ],
- "outputs": [
- {
- "name": "CONDITIONING",
- "type": "CONDITIONING",
- "links": [
- 60
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "Node name for S&R": "CLIPTextEncode"
- },
+ "inputs": [],
+ "outputs": [],
+ "properties": {},
"widgets_values": [
- "shaky, glitchy, low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly, transition"
- ]
+ "Last frame conditioning, the frame size shoudl match the generated video aspect ration."
+ ],
+ "color": "#432",
+ "bgcolor": "#653"
},
{
- "id": 16,
- "type": "CLIPLoader",
+ "id": 27,
+ "type": "LTXVAddGuide",
"pos": [
- 625.4815063476562,
- 1078.2762451171875
+ 1491.158935546875,
+ 1197.301025390625
],
"size": [
315,
- 82
+ 162
],
"flags": {},
- "order": 2,
+ "order": 16,
"mode": 0,
- "inputs": [],
+ "inputs": [
+ {
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 66
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 67
+ },
+ {
+ "name": "vae",
+ "localized_name": "vae",
+ "type": "VAE",
+ "link": 75
+ },
+ {
+ "name": "latent",
+ "localized_name": "latent",
+ "type": "LATENT",
+ "link": 68
+ },
+ {
+ "name": "image",
+ "localized_name": "image",
+ "type": "IMAGE",
+ "link": 69
+ }
+ ],
"outputs": [
{
- "name": "CLIP",
- "type": "CLIP",
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
"links": [
- 18,
- 19
+ 70
],
"slot_index": 0
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "links": [
+ 71
+ ],
+ "slot_index": 1
+ },
+ {
+ "name": "latent",
+ "localized_name": "latent",
+ "type": "LATENT",
+ "links": [
+ 72,
+ 78
+ ],
+ "slot_index": 2
}
],
"properties": {
- "Node name for S&R": "CLIPLoader"
+ "Node name for S&R": "LTXVAddGuide"
},
"widgets_values": [
- "t5xxl_fp16.safetensors",
- "ltxv",
- "default"
+ 120,
+ 1
]
},
{
- "id": 31,
- "type": "LTXVScheduler",
+ "id": 10,
+ "type": "LTXVConditioning",
"pos": [
- 1889.956787109375,
- 1356.0611572265625
+ 1881.857666015625,
+ 1204.921142578125
],
"size": [
315,
- 154
+ 78
],
"flags": {},
- "order": 20,
+ "order": 17,
"mode": 0,
"inputs": [
{
- "name": "latent",
- "type": "LATENT",
- "shape": 7,
- "link": 78
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 70
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 71
}
],
"outputs": [
{
- "name": "SIGMAS",
- "type": "SIGMAS",
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
"links": [
- 79
+ 87,
+ 90
],
"slot_index": 0
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "links": [
+ 88,
+ 91
+ ],
+ "slot_index": 1
}
],
"properties": {
- "Node name for S&R": "LTXVScheduler"
+ "Node name for S&R": "LTXVConditioning"
},
"widgets_values": [
- 20,
- 2.05,
- 0.95,
- true,
- 0.1
+ 25
]
},
{
- "id": 12,
- "type": "CheckpointLoaderSimple",
+ "id": 38,
+ "type": "STGGuiderAdvanced",
"pos": [
- 1080.441650390625,
- 1033.5550537109375
+ 1886.081787109375,
+ 1587.580322265625
],
"size": [
- 315,
- 98
+ 327.5999755859375,
+ 242
],
"flags": {},
- "order": 3,
+ "order": 19,
"mode": 0,
- "inputs": [],
- "outputs": [
+ "inputs": [
{
- "name": "MODEL",
+ "name": "model",
+ "localized_name": "model",
"type": "MODEL",
- "links": [
- 9
- ],
- "slot_index": 0
+ "link": 92
},
{
- "name": "CLIP",
- "type": "CLIP",
- "links": null
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 90
},
{
- "name": "VAE",
- "type": "VAE",
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 91
+ }
+ ],
+ "outputs": [
+ {
+ "name": "GUIDER",
+ "localized_name": "GUIDER",
+ "type": "GUIDER",
"links": [
- 73,
- 74,
- 75
+ 93
],
- "slot_index": 2
+ "slot_index": 0
}
],
"properties": {
- "Node name for S&R": "CheckpointLoaderSimple"
+ "Node name for S&R": "STGGuiderAdvanced"
},
"widgets_values": [
- "ltx-video-2b-v0.9.5.safetensors"
+ 0.996,
+ true,
+ "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "4, 4, 4, 4, 1, 1",
+ "2, 2, 2, 2, 1, 0",
+ "1, 1, 1, 1, 1, 1",
+ "[14], [14], [14], [14], [14], [14]"
]
},
{
- "id": 32,
- "type": "Note",
- "pos": [
- 2258.57373046875,
- 1423.3946533203125
- ],
- "size": [
- 291.3345947265625,
- 73.75885772705078
- ],
- "flags": {},
- "order": 4,
- "mode": 0,
- "inputs": [],
- "outputs": [],
- "properties": {},
- "widgets_values": [
- "For faster running times you can try setting cfg to 1."
- ],
- "color": "#432",
- "bgcolor": "#653"
- },
- {
- "id": 27,
- "type": "LTXVAddGuide",
+ "id": 7,
+ "type": "SamplerCustomAdvanced",
"pos": [
- 1493.655517578125,
- 1201.2786865234375
+ 2242.384521484375,
+ 1192.1341552734375
],
"size": [
- 315,
- 162
+ 355.20001220703125,
+ 106
],
"flags": {},
- "order": 18,
+ "order": 20,
"mode": 0,
"inputs": [
{
- "name": "positive",
- "type": "CONDITIONING",
- "link": 66
+ "name": "noise",
+ "localized_name": "noise",
+ "type": "NOISE",
+ "link": 3
},
{
- "name": "negative",
- "type": "CONDITIONING",
- "link": 67
+ "name": "guider",
+ "localized_name": "guider",
+ "type": "GUIDER",
+ "link": 93
},
{
- "name": "vae",
- "type": "VAE",
- "link": 75
+ "name": "sampler",
+ "localized_name": "sampler",
+ "type": "SAMPLER",
+ "link": 4
},
{
- "name": "latent",
- "type": "LATENT",
- "link": 68
+ "name": "sigmas",
+ "localized_name": "sigmas",
+ "type": "SIGMAS",
+ "link": 79
},
{
- "name": "image",
- "type": "IMAGE",
- "link": 69
+ "name": "latent_image",
+ "localized_name": "latent_image",
+ "type": "LATENT",
+ "link": 72
}
],
"outputs": [
{
- "name": "positive",
- "type": "CONDITIONING",
+ "name": "output",
+ "localized_name": "output",
+ "type": "LATENT",
"links": [
- 70
+ 85
],
"slot_index": 0
},
{
- "name": "negative",
- "type": "CONDITIONING",
- "links": [
- 71
- ],
- "slot_index": 1
- },
- {
- "name": "latent",
+ "name": "denoised_output",
+ "localized_name": "denoised_output",
"type": "LATENT",
- "links": [
- 72,
- 78
- ],
- "slot_index": 2
+ "links": null
}
],
"properties": {
- "Node name for S&R": "LTXVAddGuide"
+ "Node name for S&R": "SamplerCustomAdvanced"
},
- "widgets_values": [
- 120,
- 1
- ]
+ "widgets_values": []
},
{
- "id": 14,
- "type": "CLIPTextEncode",
+ "id": 37,
+ "type": "LTXVCropGuides",
"pos": [
- 617.918701171875,
- 1220.501708984375
+ 2292.63720703125,
+ 1359.763427734375
],
"size": [
- 400,
- 200
+ 216.59999084472656,
+ 66
],
"flags": {},
- "order": 10,
+ "order": 21,
"mode": 0,
"inputs": [
{
- "name": "clip",
- "type": "CLIP",
- "link": 18
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 87
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 88
+ },
+ {
+ "name": "latent",
+ "localized_name": "latent",
+ "type": "LATENT",
+ "link": 85
}
],
"outputs": [
{
- "name": "CONDITIONING",
+ "name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
+ "links": null
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "links": null
+ },
+ {
+ "name": "latent",
+ "localized_name": "latent",
+ "type": "LATENT",
"links": [
- 59
+ 86
],
- "slot_index": 0
+ "slot_index": 2
}
],
"properties": {
- "Node name for S&R": "CLIPTextEncode"
+ "Node name for S&R": "LTXVCropGuides"
},
- "widgets_values": [
- "A donkey gracefully rides a surfboard, going from a fiery volcanic wave to a frozen, icy landscape. Initially, the dark brown donkey stands on a surfboard amidst a wave of lava, the intense orange flames illuminating its silhouette. The camera maintains a medium shot, slightly angled, capturing the donkey's determined expression. The scene shifts smoothly to a wide shot of the donkey on the same surfboard, now gliding across a frozen lake with scattered ice floes and snow-covered rocks. The lighting changes to a bright, clear winter day. The donkey's expression remains calm; its movement is smooth and controlled. The video is computer-generated imagery.\n\n\n\n\n\n\n"
- ]
+ "widgets_values": []
},
{
- "id": 2,
- "type": "LoadImage",
+ "id": 17,
+ "type": "VAEDecode",
"pos": [
- 1091.1929931640625,
- 1840.934814453125
+ 2298.669677734375,
+ 1490.08056640625
],
"size": [
- 315,
- 314
+ 210,
+ 46
],
"flags": {},
- "order": 5,
+ "order": 22,
"mode": 0,
- "inputs": [],
- "outputs": [
+ "inputs": [
{
- "name": "IMAGE",
- "type": "IMAGE",
- "links": [
- 80
- ],
- "slot_index": 0
+ "name": "samples",
+ "localized_name": "samples",
+ "type": "LATENT",
+ "link": 86
},
{
- "name": "MASK",
- "type": "MASK",
- "links": null
+ "name": "vae",
+ "localized_name": "vae",
+ "type": "VAE",
+ "link": 73
}
],
- "properties": {
- "Node name for S&R": "LoadImage"
- },
- "widgets_values": [
- "start.jpg",
- "image"
- ]
- },
- {
- "id": 1,
- "type": "LoadImage",
- "pos": [
- 1502.8370361328125,
- 1844.201171875
- ],
- "size": [
- 315,
- 314
- ],
- "flags": {},
- "order": 6,
- "mode": 0,
- "inputs": [],
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
- 82
+ 84
],
"slot_index": 0
- },
- {
- "name": "MASK",
- "type": "MASK",
- "links": null
}
],
"properties": {
- "Node name for S&R": "LoadImage"
+ "Node name for S&R": "VAEDecode"
},
- "widgets_values": [
- "end.jpg",
- "image"
- ]
+ "widgets_values": []
},
{
- "id": 34,
- "type": "ImageResizeKJ",
+ "id": 36,
+ "type": "VHS_VideoCombine",
"pos": [
- 1092.6722412109375,
- 1530.77978515625
+ 2296.265380859375,
+ 1605.352783203125
],
"size": [
315,
- 266
+ 520.6666870117188
],
"flags": {},
- "order": 13,
+ "order": 23,
"mode": 0,
"inputs": [
{
- "name": "image",
+ "name": "images",
+ "localized_name": "images",
"type": "IMAGE",
- "link": 80
+ "shape": 7,
+ "link": 84
},
{
- "name": "get_image_size",
- "type": "IMAGE",
+ "name": "audio",
+ "localized_name": "audio",
+ "type": "AUDIO",
"shape": 7,
"link": null
},
{
- "name": "width_input",
- "type": "INT",
+ "name": "meta_batch",
+ "localized_name": "meta_batch",
+ "type": "VHS_BatchManager",
"shape": 7,
- "widget": {
- "name": "width_input"
- },
"link": null
},
{
- "name": "height_input",
- "type": "INT",
+ "name": "vae",
+ "localized_name": "vae",
+ "type": "VAE",
"shape": 7,
- "widget": {
- "name": "height_input"
- },
"link": null
}
],
"outputs": [
{
- "name": "IMAGE",
- "type": "IMAGE",
- "links": [
- 81
- ],
- "slot_index": 0
- },
- {
- "name": "width",
- "type": "INT",
- "links": null
- },
- {
- "name": "height",
- "type": "INT",
+ "name": "Filenames",
+ "localized_name": "Filenames",
+ "type": "VHS_FILENAMES",
"links": null
}
],
"properties": {
- "Node name for S&R": "ImageResizeKJ"
+ "Node name for S&R": "VHS_VideoCombine"
},
- "widgets_values": [
- 768,
- 512,
- "nearest-exact",
- false,
- 2,
- 0,
- 0,
- "center"
- ]
+ "widgets_values": {
+ "frame_rate": 25,
+ "loop_count": 0,
+ "filename_prefix": "ltxv",
+ "format": "video/h264-mp4",
+ "pix_fmt": "yuv420p",
+ "crf": 19,
+ "save_metadata": true,
+ "pingpong": false,
+ "save_output": true,
+ "videopreview": {
+ "hidden": false,
+ "paused": false,
+ "params": {
+ "filename": "ltxv_00121.mp4",
+ "subfolder": "",
+ "type": "output",
+ "format": "video/h264-mp4",
+ "frame_rate": 25
+ }
+ }
+ }
},
{
- "id": 35,
+ "id": 34,
"type": "ImageResizeKJ",
"pos": [
- 1499.1163330078125,
- 1531.5040283203125
+ 1090.9002685546875,
+ 1524.03857421875
],
"size": [
315,
- 266
+ 286
],
"flags": {},
- "order": 14,
+ "order": 9,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
- "link": 82
+ "link": 80
},
{
"name": "get_image_size",
+ "localized_name": "get_image_size",
"type": "IMAGE",
"shape": 7,
"link": null
@@ -854,6 +911,10 @@
"name": "width_input",
"type": "INT",
"shape": 7,
+ "pos": [
+ 10,
+ 196
+ ],
"widget": {
"name": "width_input"
},
@@ -863,6 +924,10 @@
"name": "height_input",
"type": "INT",
"shape": 7,
+ "pos": [
+ 10,
+ 220
+ ],
"widget": {
"name": "height_input"
},
@@ -872,19 +937,22 @@
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
- 83
+ 81
],
"slot_index": 0
},
{
"name": "width",
+ "localized_name": "width",
"type": "INT",
"links": null
},
{
"name": "height",
+ "localized_name": "height",
"type": "INT",
"links": null
}
@@ -904,291 +972,176 @@
]
},
{
- "id": 28,
- "type": "Note",
- "pos": [
- 1095.3414306640625,
- 2204.1455078125
- ],
- "size": [
- 314.0157470703125,
- 91.49161529541016
- ],
- "flags": {},
- "order": 7,
- "mode": 0,
- "inputs": [],
- "outputs": [],
- "properties": {},
- "widgets_values": [
- "First frame conditioning, the frame size shoudl match the generated video aspect ration."
- ],
- "color": "#432",
- "bgcolor": "#653"
- },
- {
- "id": 29,
- "type": "Note",
- "pos": [
- 1510.7900390625,
- 2206.3818359375
- ],
- "size": [
- 304.03729248046875,
- 80.94778442382812
- ],
- "flags": {},
- "order": 8,
- "mode": 0,
- "inputs": [],
- "outputs": [],
- "properties": {},
- "widgets_values": [
- "Last frame conditioning, the frame size shoudl match the generated video aspect ration."
- ],
- "color": "#432",
- "bgcolor": "#653"
- },
- {
- "id": 10,
- "type": "LTXVConditioning",
+ "id": 13,
+ "type": "EmptyLTXVLatentVideo",
"pos": [
- 1881.857666015625,
- 1204.921142578125
+ 658.8583374023438,
+ 1901.9586181640625
],
"size": [
315,
- 78
+ 130
],
"flags": {},
- "order": 19,
+ "order": 4,
"mode": 0,
- "inputs": [
- {
- "name": "positive",
- "type": "CONDITIONING",
- "link": 70
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "link": 71
- }
- ],
+ "inputs": [],
"outputs": [
{
- "name": "positive",
- "type": "CONDITIONING",
+ "name": "LATENT",
+ "localized_name": "LATENT",
+ "type": "LATENT",
"links": [
- 7,
- 87
+ 62
],
"slot_index": 0
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "links": [
- 8,
- 88
- ],
- "slot_index": 1
}
],
"properties": {
- "Node name for S&R": "LTXVConditioning"
+ "Node name for S&R": "EmptyLTXVLatentVideo"
},
"widgets_values": [
- 25
+ 768,
+ 512,
+ 121,
+ 1
]
},
{
- "id": 36,
- "type": "VHS_VideoCombine",
+ "id": 15,
+ "type": "CLIPTextEncode",
"pos": [
- 2649.404052734375,
- 1612.633056640625
+ 629.337158203125,
+ 1725.8612060546875
],
"size": [
- 315,
- 520.6666870117188
+ 392.1722717285156,
+ 119.6710433959961
],
"flags": {},
- "order": 25,
+ "order": 12,
"mode": 0,
"inputs": [
{
- "name": "images",
- "type": "IMAGE",
- "shape": 7,
- "link": 84
- },
- {
- "name": "audio",
- "type": "AUDIO",
- "shape": 7,
- "link": null
- },
- {
- "name": "meta_batch",
- "type": "VHS_BatchManager",
- "shape": 7,
- "link": null
- },
- {
- "name": "vae",
- "type": "VAE",
- "shape": 7,
- "link": null
+ "name": "clip",
+ "localized_name": "clip",
+ "type": "CLIP",
+ "link": 19
}
],
"outputs": [
{
- "name": "Filenames",
- "type": "VHS_FILENAMES",
- "links": null
+ "name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
+ "type": "CONDITIONING",
+ "links": [
+ 60
+ ],
+ "slot_index": 0
}
],
"properties": {
- "Node name for S&R": "VHS_VideoCombine"
+ "Node name for S&R": "CLIPTextEncode"
},
- "widgets_values": {
- "frame_rate": 25,
- "loop_count": 0,
- "filename_prefix": "ltxv",
- "format": "video/h264-mp4",
- "pix_fmt": "yuv420p",
- "crf": 19,
- "save_metadata": true,
- "pingpong": false,
- "save_output": true,
- "videopreview": {
- "hidden": false,
- "paused": false,
- "params": {
- "filename": "ltxv_00132.mp4",
- "subfolder": "",
- "type": "output",
- "format": "video/h264-mp4",
- "frame_rate": 25
- }
- }
- }
+ "widgets_values": [
+ "shaky, glitchy, low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly, transition"
+ ]
},
{
- "id": 17,
- "type": "VAEDecode",
+ "id": 14,
+ "type": "CLIPTextEncode",
"pos": [
- 2669.6435546875,
- 1471.837646484375
+ 624.9329833984375,
+ 1471.5355224609375
],
"size": [
- 210,
- 46
+ 400,
+ 200
],
"flags": {},
- "order": 24,
+ "order": 11,
"mode": 0,
"inputs": [
{
- "name": "samples",
- "type": "LATENT",
- "link": 86
- },
- {
- "name": "vae",
- "type": "VAE",
- "link": 73
+ "name": "clip",
+ "localized_name": "clip",
+ "type": "CLIP",
+ "link": 18
}
],
"outputs": [
{
- "name": "IMAGE",
- "type": "IMAGE",
+ "name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
+ "type": "CONDITIONING",
"links": [
- 84
+ 59
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "VAEDecode"
+ "Node name for S&R": "CLIPTextEncode"
},
- "widgets_values": []
+ "widgets_values": [
+ "A donkey gracefully rides a surfboard, going from a fiery volcanic wave to a frozen, icy landscape. Initially, the dark brown donkey stands on a surfboard amidst a wave of lava, the intense orange flames illuminating its silhouette. The camera maintains a medium shot, slightly angled, capturing the donkey's determined expression. The scene shifts smoothly to a wide shot of the donkey on the same surfboard, now gliding across a frozen lake with scattered ice floes and snow-covered rocks. The lighting changes to a bright, clear winter day. The donkey's expression remains calm; its movement is smooth and controlled. The video is computer-generated imagery.\n\n\n\n\n\n\n"
+ ]
},
{
- "id": 37,
- "type": "LTXVCropGuides",
+ "id": 16,
+ "type": "CLIPLoader",
"pos": [
- 2688.18603515625,
- 1336.5384521484375
+ 657.5050048828125,
+ 1323.1318359375
],
"size": [
- 216.59999084472656,
- 66
+ 315,
+ 98
],
"flags": {},
- "order": 23,
+ "order": 5,
"mode": 0,
- "inputs": [
- {
- "name": "positive",
- "type": "CONDITIONING",
- "link": 87
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "link": 88
- },
- {
- "name": "latent",
- "type": "LATENT",
- "link": 85
- }
- ],
+ "inputs": [],
"outputs": [
{
- "name": "positive",
- "type": "CONDITIONING",
- "links": null
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "links": null
- },
- {
- "name": "latent",
- "type": "LATENT",
+ "name": "CLIP",
+ "localized_name": "CLIP",
+ "type": "CLIP",
"links": [
- 86
+ 18,
+ 19
],
- "slot_index": 2
+ "slot_index": 0
}
],
"properties": {
- "Node name for S&R": "LTXVCropGuides"
- }
+ "Node name for S&R": "CLIPLoader"
+ },
+ "widgets_values": [
+ "t5xxl_fp16.safetensors",
+ "ltxv",
+ "default"
+ ]
},
{
"id": 8,
"type": "RandomNoise",
"pos": [
- 1850.8992919921875,
- 1055.1177978515625
+ 1891.2080078125,
+ 1895.3349609375
],
"size": [
315,
82
],
"flags": {},
- "order": 9,
+ "order": 6,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "NOISE",
+ "localized_name": "NOISE",
"type": "NOISE",
"links": [
3
@@ -1203,17 +1156,92 @@
5,
"fixed"
]
+ },
+ {
+ "id": 9,
+ "type": "KSamplerSelect",
+ "pos": [
+ 1896.1917724609375,
+ 2035.409423828125
+ ],
+ "size": [
+ 315,
+ 58
+ ],
+ "flags": {},
+ "order": 7,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "SAMPLER",
+ "localized_name": "SAMPLER",
+ "type": "SAMPLER",
+ "links": [
+ 4
+ ],
+ "slot_index": 0
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "KSamplerSelect"
+ },
+ "widgets_values": [
+ "gradient_estimation"
+ ]
+ },
+ {
+ "id": 12,
+ "type": "CheckpointLoaderSimple",
+ "pos": [
+ 655.3289794921875,
+ 1178.0711669921875
+ ],
+ "size": [
+ 315,
+ 98
+ ],
+ "flags": {},
+ "order": 8,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "MODEL",
+ "localized_name": "MODEL",
+ "type": "MODEL",
+ "links": [
+ 92
+ ],
+ "slot_index": 0
+ },
+ {
+ "name": "CLIP",
+ "localized_name": "CLIP",
+ "type": "CLIP",
+ "links": null
+ },
+ {
+ "name": "VAE",
+ "localized_name": "VAE",
+ "type": "VAE",
+ "links": [
+ 73,
+ 74,
+ 75
+ ],
+ "slot_index": 2
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "CheckpointLoaderSimple"
+ },
+ "widgets_values": [
+ "ltxv-2b-0.9.6-dev-04-25.safetensors"
+ ]
}
],
"links": [
- [
- 2,
- 5,
- 0,
- 7,
- 1,
- "GUIDER"
- ],
[
3,
8,
@@ -1230,38 +1258,6 @@
2,
"SAMPLER"
],
- [
- 7,
- 10,
- 0,
- 5,
- 1,
- "CONDITIONING"
- ],
- [
- 8,
- 10,
- 1,
- 5,
- 2,
- "CONDITIONING"
- ],
- [
- 9,
- 12,
- 0,
- 11,
- 0,
- "MODEL"
- ],
- [
- 10,
- 11,
- 0,
- 5,
- 0,
- "MODEL"
- ],
[
18,
16,
@@ -1477,20 +1473,52 @@
37,
1,
"CONDITIONING"
+ ],
+ [
+ 90,
+ 10,
+ 0,
+ 38,
+ 1,
+ "CONDITIONING"
+ ],
+ [
+ 91,
+ 10,
+ 1,
+ 38,
+ 2,
+ "CONDITIONING"
+ ],
+ [
+ 92,
+ 12,
+ 0,
+ 38,
+ 0,
+ "MODEL"
+ ],
+ [
+ 93,
+ 38,
+ 0,
+ 7,
+ 1,
+ "GUIDER"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
- "scale": 0.6538676505623344,
+ "scale": 0.6303940863128502,
"offset": [
- -885.0991685047312,
- -1012.4869764971703
+ -355.757443222888,
+ -1116.3282097448212
]
},
"node_versions": {
- "comfy-core": "0.3.18"
+ "comfy-core": "0.3.27"
},
"prompt": {
"1": {
@@ -1513,29 +1541,6 @@
"title": "Load Image"
}
},
- "5": {
- "inputs": {
- "cfg": 3,
- "stg": 1,
- "rescale": 0.75,
- "model": [
- "11",
- 0
- ],
- "positive": [
- "10",
- 0
- ],
- "negative": [
- "10",
- 1
- ]
- },
- "class_type": "STGGuider",
- "_meta": {
- "title": "🅛🅣🅧 STG Guider"
- }
- },
"7": {
"inputs": {
"noise": [
@@ -1543,7 +1548,7 @@
0
],
"guider": [
- "5",
+ "38",
0
],
"sampler": [
@@ -1599,22 +1604,9 @@
"title": "LTXVConditioning"
}
},
- "11": {
- "inputs": {
- "block_indices": "14",
- "model": [
- "12",
- 0
- ]
- },
- "class_type": "LTXVApplySTG",
- "_meta": {
- "title": "🅛🅣🅧 LTXV Apply STG"
- }
- },
"12": {
"inputs": {
- "ckpt_name": "ltx-video-2b-v0.9.5.safetensors"
+ "ckpt_name": "ltxv-2b-0.9.6-dev-04-25.safetensors"
},
"class_type": "CheckpointLoaderSimple",
"_meta": {
@@ -1865,9 +1857,36 @@
"_meta": {
"title": "LTXVCropGuides"
}
+ },
+ "38": {
+ "inputs": {
+ "skip_steps_sigma_threshold": 0.996,
+ "cfg_star_rescale": true,
+ "sigmas": "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "cfg_values": "4, 4, 4, 4, 1, 1",
+ "stg_scale_values": "2, 2, 2, 2, 1, 0",
+ "stg_rescale_values": "1, 1, 1, 1, 1, 1",
+ "stg_layers_indices": "[14], [14], [14], [14], [14], [14]",
+ "model": [
+ "12",
+ 0
+ ],
+ "positive": [
+ "10",
+ 0
+ ],
+ "negative": [
+ "10",
+ 1
+ ]
+ },
+ "class_type": "STGGuiderAdvanced",
+ "_meta": {
+ "title": "🅛🅣🅧 STG Guider Advanced"
+ }
}
},
- "comfy_fork_version": "update_nodes@db0f122d"
+ "comfy_fork_version": "feature/memory_prof@f7e64946"
},
"version": 0.4
}
diff --git a/assets/ltxvideo-frame-interpolation.png b/assets/ltxvideo-frame-interpolation.png
index eb3a14e..f56e410 100644
--- a/assets/ltxvideo-frame-interpolation.png
+++ b/assets/ltxvideo-frame-interpolation.png
@@ -1,3 +1,3 @@
version https://git-lfs.github.com/spec/v1
-oid sha256:97ea4a9a45004726ae84971bef44578270a7be986b1fce533cc8d3ec1a213fdd
-size 1390379
+oid sha256:77d9f98afa1a5711fade39a6fa08b9c3eab8c439c3727f522bae639bee11cf0a
+size 1380444
diff --git a/assets/ltxvideo-i2v-distilled.json b/assets/ltxvideo-i2v-distilled.json
new file mode 100644
index 0000000..2260170
--- /dev/null
+++ b/assets/ltxvideo-i2v-distilled.json
@@ -0,0 +1,1009 @@
+{
+ "id": "d8be2e45-0fd2-4f4a-8065-469fdf4bc018",
+ "revision": 0,
+ "last_node_id": 109,
+ "last_link_id": 268,
+ "nodes": [
+ {
+ "id": 38,
+ "type": "CLIPLoader",
+ "pos": [
+ 60,
+ 190
+ ],
+ "size": [
+ 315,
+ 98
+ ],
+ "flags": {},
+ "order": 0,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "CLIP",
+ "type": "CLIP",
+ "slot_index": 0,
+ "links": [
+ 74,
+ 75
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "CLIPLoader"
+ },
+ "widgets_values": [
+ "t5xxl_fp16.safetensors",
+ "ltxv",
+ "default"
+ ]
+ },
+ {
+ "id": 8,
+ "type": "VAEDecode",
+ "pos": [
+ 1800,
+ 430
+ ],
+ "size": [
+ 210,
+ 46
+ ],
+ "flags": {},
+ "order": 16,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "samples",
+ "type": "LATENT",
+ "link": 255
+ },
+ {
+ "name": "vae",
+ "type": "VAE",
+ "link": 87
+ }
+ ],
+ "outputs": [
+ {
+ "name": "IMAGE",
+ "type": "IMAGE",
+ "slot_index": 0,
+ "links": [
+ 261
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "VAEDecode"
+ },
+ "widgets_values": []
+ },
+ {
+ "id": 103,
+ "type": "VHS_VideoCombine",
+ "pos": [
+ 1789.6087646484375,
+ 548.7745361328125
+ ],
+ "size": [
+ 315,
+ 545
+ ],
+ "flags": {},
+ "order": 17,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "images",
+ "shape": 7,
+ "type": "IMAGE",
+ "link": 261
+ },
+ {
+ "name": "audio",
+ "shape": 7,
+ "type": "AUDIO",
+ "link": null
+ },
+ {
+ "name": "meta_batch",
+ "shape": 7,
+ "type": "VHS_BatchManager",
+ "link": null
+ },
+ {
+ "name": "vae",
+ "shape": 7,
+ "type": "VAE",
+ "link": null
+ }
+ ],
+ "outputs": [
+ {
+ "name": "Filenames",
+ "type": "VHS_FILENAMES",
+ "links": null
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfyui-videohelpersuite",
+ "ver": "972c87da577b47211c4e9aeed30dc38c7bae607f",
+ "Node name for S&R": "VHS_VideoCombine"
+ },
+ "widgets_values": {
+ "frame_rate": 24,
+ "loop_count": 0,
+ "filename_prefix": "ltxv",
+ "format": "video/h264-mp4",
+ "pix_fmt": "yuv420p",
+ "crf": 20,
+ "save_metadata": false,
+ "trim_to_audio": false,
+ "pingpong": false,
+ "save_output": true,
+ "videopreview": {
+ "hidden": false,
+ "paused": false,
+ "params": {
+ "filename": "ltxv_00058.mp4",
+ "subfolder": "",
+ "type": "output",
+ "format": "video/h264-mp4",
+ "frame_rate": 24,
+ "workflow": "ltxv_00058.png",
+ "fullpath": "C:\\Users\\Zeev\\Projects\\ComfyUI\\output\\ltxv_00058.mp4"
+ }
+ }
+ }
+ },
+ {
+ "id": 82,
+ "type": "LTXVPreprocess",
+ "pos": [
+ 488.92791748046875,
+ 629.9364624023438
+ ],
+ "size": [
+ 275.9266662597656,
+ 58
+ ],
+ "flags": {},
+ "order": 10,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "image",
+ "type": "IMAGE",
+ "link": 226
+ }
+ ],
+ "outputs": [
+ {
+ "name": "output_image",
+ "type": "IMAGE",
+ "slot_index": 0,
+ "links": [
+ 248
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "LTXVPreprocess"
+ },
+ "widgets_values": [
+ 38
+ ]
+ },
+ {
+ "id": 7,
+ "type": "CLIPTextEncode",
+ "pos": [
+ 420,
+ 390
+ ],
+ "size": [
+ 425.27801513671875,
+ 180.6060791015625
+ ],
+ "flags": {},
+ "order": 9,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "clip",
+ "type": "CLIP",
+ "link": 75
+ }
+ ],
+ "outputs": [
+ {
+ "name": "CONDITIONING",
+ "type": "CONDITIONING",
+ "slot_index": 0,
+ "links": [
+ 240
+ ]
+ }
+ ],
+ "title": "CLIP Text Encode (Negative Prompt)",
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "CLIPTextEncode"
+ },
+ "widgets_values": [
+ "low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly"
+ ],
+ "color": "#322",
+ "bgcolor": "#533"
+ },
+ {
+ "id": 78,
+ "type": "LoadImage",
+ "pos": [
+ 40,
+ 630
+ ],
+ "size": [
+ 385.15606689453125,
+ 333.3305358886719
+ ],
+ "flags": {},
+ "order": 1,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "IMAGE",
+ "type": "IMAGE",
+ "slot_index": 0,
+ "links": [
+ 226
+ ]
+ },
+ {
+ "name": "MASK",
+ "type": "MASK",
+ "links": null
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "LoadImage"
+ },
+ "widgets_values": [
+ "fox.jpg",
+ "image",
+ ""
+ ]
+ },
+ {
+ "id": 6,
+ "type": "CLIPTextEncode",
+ "pos": [
+ 420,
+ 180
+ ],
+ "size": [
+ 422.84503173828125,
+ 164.31304931640625
+ ],
+ "flags": {},
+ "order": 8,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "clip",
+ "type": "CLIP",
+ "link": 74
+ }
+ ],
+ "outputs": [
+ {
+ "name": "CONDITIONING",
+ "type": "CONDITIONING",
+ "slot_index": 0,
+ "links": [
+ 239
+ ]
+ }
+ ],
+ "title": "CLIP Text Encode (Positive Prompt)",
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "CLIPTextEncode"
+ },
+ "widgets_values": [
+ "A red fox moving gracefully, its russet coat vibrant against the white landscape, leaving perfect star-shaped prints behind as steam rises from its breath in the crisp winter air. The scene is wrapped in snow-muffled silence, broken only by the gentle murmur of water still flowing beneath the ice."
+ ],
+ "color": "#232",
+ "bgcolor": "#353"
+ },
+ {
+ "id": 76,
+ "type": "Note",
+ "pos": [
+ 36.85359573364258,
+ 360.48809814453125
+ ],
+ "size": [
+ 360,
+ 200
+ ],
+ "flags": {},
+ "order": 2,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [],
+ "properties": {},
+ "widgets_values": [
+ "While LTXV-2b model prefers long descriptive prompts, this version supports experimentation with broader prompting styles."
+ ],
+ "color": "#432",
+ "bgcolor": "#653"
+ },
+ {
+ "id": 69,
+ "type": "LTXVConditioning",
+ "pos": [
+ 1183.358642578125,
+ 292.5882873535156
+ ],
+ "size": [
+ 223.8660125732422,
+ 78
+ ],
+ "flags": {},
+ "order": 13,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "positive",
+ "type": "CONDITIONING",
+ "link": 245
+ },
+ {
+ "name": "negative",
+ "type": "CONDITIONING",
+ "link": 246
+ }
+ ],
+ "outputs": [
+ {
+ "name": "positive",
+ "type": "CONDITIONING",
+ "slot_index": 0,
+ "links": [
+ 256
+ ]
+ },
+ {
+ "name": "negative",
+ "type": "CONDITIONING",
+ "slot_index": 1,
+ "links": [
+ 257
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "LTXVConditioning"
+ },
+ "widgets_values": [
+ 24.000000000000004
+ ]
+ },
+ {
+ "id": 101,
+ "type": "SamplerCustomAdvanced",
+ "pos": [
+ 1780,
+ 270
+ ],
+ "size": [
+ 236.8000030517578,
+ 106
+ ],
+ "flags": {},
+ "order": 15,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "noise",
+ "type": "NOISE",
+ "link": 260
+ },
+ {
+ "name": "guider",
+ "type": "GUIDER",
+ "link": 252
+ },
+ {
+ "name": "sampler",
+ "type": "SAMPLER",
+ "link": 253
+ },
+ {
+ "name": "sigmas",
+ "type": "SIGMAS",
+ "link": 268
+ },
+ {
+ "name": "latent_image",
+ "type": "LATENT",
+ "link": 259
+ }
+ ],
+ "outputs": [
+ {
+ "name": "output",
+ "type": "LATENT",
+ "slot_index": 0,
+ "links": []
+ },
+ {
+ "name": "denoised_output",
+ "type": "LATENT",
+ "slot_index": 1,
+ "links": [
+ 255
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.15",
+ "Node name for S&R": "SamplerCustomAdvanced"
+ },
+ "widgets_values": []
+ },
+ {
+ "id": 95,
+ "type": "LTXVImgToVideo",
+ "pos": [
+ 900,
+ 290
+ ],
+ "size": [
+ 210,
+ 190
+ ],
+ "flags": {},
+ "order": 12,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "positive",
+ "type": "CONDITIONING",
+ "link": 239
+ },
+ {
+ "name": "negative",
+ "type": "CONDITIONING",
+ "link": 240
+ },
+ {
+ "name": "vae",
+ "type": "VAE",
+ "link": 250
+ },
+ {
+ "name": "image",
+ "type": "IMAGE",
+ "link": 248
+ }
+ ],
+ "outputs": [
+ {
+ "name": "positive",
+ "type": "CONDITIONING",
+ "slot_index": 0,
+ "links": [
+ 245
+ ]
+ },
+ {
+ "name": "negative",
+ "type": "CONDITIONING",
+ "slot_index": 1,
+ "links": [
+ 246
+ ]
+ },
+ {
+ "name": "latent",
+ "type": "LATENT",
+ "slot_index": 2,
+ "links": [
+ 259
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "LTXVImgToVideo"
+ },
+ "widgets_values": [
+ 768,
+ 512,
+ 97,
+ 1
+ ]
+ },
+ {
+ "id": 100,
+ "type": "StringToFloatList",
+ "pos": [
+ 1116.0089111328125,
+ 604.5989379882812
+ ],
+ "size": [
+ 395.74224853515625,
+ 88
+ ],
+ "flags": {},
+ "order": 3,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "FLOAT",
+ "type": "FLOAT",
+ "links": [
+ 251
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfyui-kjnodes",
+ "ver": "0addfc6101f7a834c7fb6e0a1b26529360ab5350",
+ "Node name for S&R": "StringToFloatList"
+ },
+ "widgets_values": [
+ "1.0000, 0.9937, 0.9875, 0.9812, 0.9750, 0.9094, 0.7250, 0.4219, 0.0"
+ ],
+ "color": "#223",
+ "bgcolor": "#335"
+ },
+ {
+ "id": 105,
+ "type": "Note",
+ "pos": [
+ 1126.7423095703125,
+ 424.2294616699219
+ ],
+ "size": [
+ 335.8657531738281,
+ 106.6832046508789
+ ],
+ "flags": {},
+ "order": 4,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [],
+ "properties": {},
+ "widgets_values": [
+ "Distilled model expects the following sigma schedule:\n1.0000, 0.9937, 0.9875, 0.9812, 0.9750, 0.9094, 0.7250, 0.4219, 0.0\n\n\nEuler_ancestral is the recommended default sampler."
+ ],
+ "color": "#432",
+ "bgcolor": "#653"
+ },
+ {
+ "id": 97,
+ "type": "KSamplerSelect",
+ "pos": [
+ 1477.55615234375,
+ 427.04132080078125
+ ],
+ "size": [
+ 275.599365234375,
+ 58
+ ],
+ "flags": {},
+ "order": 5,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "SAMPLER",
+ "type": "SAMPLER",
+ "slot_index": 0,
+ "links": [
+ 253
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.15",
+ "Node name for S&R": "KSamplerSelect"
+ },
+ "widgets_values": [
+ "euler_ancestral"
+ ]
+ },
+ {
+ "id": 99,
+ "type": "CFGGuider",
+ "pos": [
+ 1536.1513671875,
+ 280
+ ],
+ "size": [
+ 210,
+ 98
+ ],
+ "flags": {},
+ "order": 14,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "model",
+ "type": "MODEL",
+ "link": 258
+ },
+ {
+ "name": "positive",
+ "type": "CONDITIONING",
+ "link": 256
+ },
+ {
+ "name": "negative",
+ "type": "CONDITIONING",
+ "link": 257
+ }
+ ],
+ "outputs": [
+ {
+ "name": "GUIDER",
+ "type": "GUIDER",
+ "slot_index": 0,
+ "links": [
+ 252
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.26",
+ "Node name for S&R": "CFGGuider"
+ },
+ "widgets_values": [
+ 1
+ ]
+ },
+ {
+ "id": 102,
+ "type": "RandomNoise",
+ "pos": [
+ 1524.6190185546875,
+ 138.8462371826172
+ ],
+ "size": [
+ 210,
+ 82
+ ],
+ "flags": {},
+ "order": 6,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "NOISE",
+ "type": "NOISE",
+ "links": [
+ 260
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "RandomNoise"
+ },
+ "widgets_values": [
+ 45,
+ "fixed"
+ ]
+ },
+ {
+ "id": 98,
+ "type": "FloatToSigmas",
+ "pos": [
+ 1574.70166015625,
+ 620.4625244140625
+ ],
+ "size": [
+ 210,
+ 58
+ ],
+ "flags": {
+ "collapsed": true
+ },
+ "order": 11,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "float_list",
+ "type": "FLOAT",
+ "widget": {
+ "name": "float_list"
+ },
+ "link": null
+ },
+ {
+ "name": "float_list",
+ "type": "FLOAT",
+ "widget": {
+ "name": "float_list"
+ },
+ "link": 251
+ }
+ ],
+ "outputs": [
+ {
+ "name": "SIGMAS",
+ "type": "SIGMAS",
+ "links": [
+ 268
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfyui-kjnodes",
+ "ver": "0addfc6101f7a834c7fb6e0a1b26529360ab5350",
+ "Node name for S&R": "FloatToSigmas"
+ },
+ "widgets_values": [
+ 0
+ ]
+ },
+ {
+ "id": 44,
+ "type": "CheckpointLoaderSimple",
+ "pos": [
+ 899.0839233398438,
+ 113.54173278808594
+ ],
+ "size": [
+ 315,
+ 98
+ ],
+ "flags": {},
+ "order": 7,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "MODEL",
+ "type": "MODEL",
+ "slot_index": 0,
+ "links": [
+ 258
+ ]
+ },
+ {
+ "name": "CLIP",
+ "type": "CLIP",
+ "links": null
+ },
+ {
+ "name": "VAE",
+ "type": "VAE",
+ "slot_index": 2,
+ "links": [
+ 87,
+ 250
+ ]
+ }
+ ],
+ "properties": {
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28",
+ "Node name for S&R": "CheckpointLoaderSimple"
+ },
+ "widgets_values": [
+ "ltxv-2b-0.9.6-distilled-04-25.safetensors"
+ ]
+ }
+ ],
+ "links": [
+ [
+ 74,
+ 38,
+ 0,
+ 6,
+ 0,
+ "CLIP"
+ ],
+ [
+ 75,
+ 38,
+ 0,
+ 7,
+ 0,
+ "CLIP"
+ ],
+ [
+ 87,
+ 44,
+ 2,
+ 8,
+ 1,
+ "VAE"
+ ],
+ [
+ 226,
+ 78,
+ 0,
+ 82,
+ 0,
+ "IMAGE"
+ ],
+ [
+ 239,
+ 6,
+ 0,
+ 95,
+ 0,
+ "CONDITIONING"
+ ],
+ [
+ 240,
+ 7,
+ 0,
+ 95,
+ 1,
+ "CONDITIONING"
+ ],
+ [
+ 245,
+ 95,
+ 0,
+ 69,
+ 0,
+ "CONDITIONING"
+ ],
+ [
+ 246,
+ 95,
+ 1,
+ 69,
+ 1,
+ "CONDITIONING"
+ ],
+ [
+ 248,
+ 82,
+ 0,
+ 95,
+ 3,
+ "IMAGE"
+ ],
+ [
+ 250,
+ 44,
+ 2,
+ 95,
+ 2,
+ "VAE"
+ ],
+ [
+ 251,
+ 100,
+ 0,
+ 98,
+ 1,
+ "FLOAT"
+ ],
+ [
+ 252,
+ 99,
+ 0,
+ 101,
+ 1,
+ "GUIDER"
+ ],
+ [
+ 253,
+ 97,
+ 0,
+ 101,
+ 2,
+ "SAMPLER"
+ ],
+ [
+ 255,
+ 101,
+ 1,
+ 8,
+ 0,
+ "LATENT"
+ ],
+ [
+ 256,
+ 69,
+ 0,
+ 99,
+ 1,
+ "CONDITIONING"
+ ],
+ [
+ 257,
+ 69,
+ 1,
+ 99,
+ 2,
+ "CONDITIONING"
+ ],
+ [
+ 258,
+ 44,
+ 0,
+ 99,
+ 0,
+ "MODEL"
+ ],
+ [
+ 259,
+ 95,
+ 2,
+ 101,
+ 4,
+ "LATENT"
+ ],
+ [
+ 260,
+ 102,
+ 0,
+ 101,
+ 0,
+ "NOISE"
+ ],
+ [
+ 261,
+ 8,
+ 0,
+ 103,
+ 0,
+ "IMAGE"
+ ],
+ [
+ 268,
+ 98,
+ 0,
+ 101,
+ 3,
+ "SIGMAS"
+ ]
+ ],
+ "groups": [],
+ "config": {},
+ "extra": {
+ "ds": {
+ "scale": 0.6303940863129158,
+ "offset": [
+ 627.9092558924768,
+ 478.72279841993054
+ ]
+ },
+ "VHS_latentpreview": true,
+ "VHS_latentpreviewrate": 0,
+ "VHS_MetadataImage": true,
+ "VHS_KeepIntermediate": true
+ },
+ "version": 0.4
+}
diff --git a/assets/ltxvideo-i2v-distilled.png b/assets/ltxvideo-i2v-distilled.png
new file mode 100644
index 0000000..ffaff0b
--- /dev/null
+++ b/assets/ltxvideo-i2v-distilled.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b981ed68d5f4a25e48daf31cf4cf0e156bee27b75730dd5fd213bfb805d1c8d
+size 1342913
diff --git a/assets/ltxvideo-i2v.json b/assets/ltxvideo-i2v.json
index 946f3b6..1c5f527 100644
--- a/assets/ltxvideo-i2v.json
+++ b/assets/ltxvideo-i2v.json
@@ -1,6 +1,6 @@
{
- "last_node_id": 90,
- "last_link_id": 215,
+ "last_node_id": 92,
+ "last_link_id": 224,
"nodes": [
{
"id": 81,
@@ -14,31 +14,36 @@
106
],
"flags": {},
- "order": 17,
+ "order": 16,
"mode": 0,
"inputs": [
{
"name": "noise",
+ "localized_name": "noise",
"type": "NOISE",
"link": 194
},
{
"name": "guider",
+ "localized_name": "guider",
"type": "GUIDER",
- "link": 200
+ "link": 224
},
{
"name": "sampler",
+ "localized_name": "sampler",
"type": "SAMPLER",
"link": 196
},
{
"name": "sigmas",
+ "localized_name": "sigmas",
"type": "SIGMAS",
"link": 197
},
{
"name": "latent_image",
+ "localized_name": "latent_image",
"type": "LATENT",
"link": 198
}
@@ -46,6 +51,7 @@
"outputs": [
{
"name": "output",
+ "localized_name": "output",
"type": "LATENT",
"links": [
199
@@ -54,12 +60,15 @@
},
{
"name": "denoised_output",
+ "localized_name": "denoised_output",
"type": "LATENT",
"links": null
}
],
"properties": {
- "Node name for S&R": "SamplerCustomAdvanced"
+ "Node name for S&R": "SamplerCustomAdvanced",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": []
},
@@ -72,29 +81,33 @@
],
"size": [
315,
- 190
+ 214
],
"flags": {},
- "order": 13,
+ "order": 12,
"mode": 0,
"inputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"link": 187
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"link": 188
},
{
"name": "vae",
+ "localized_name": "vae",
"type": "VAE",
"link": 213
},
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 204
}
@@ -102,6 +115,7 @@
"outputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"links": [
183
@@ -110,6 +124,7 @@
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"links": [
184
@@ -118,6 +133,7 @@
},
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"links": [
185,
@@ -127,12 +143,15 @@
}
],
"properties": {
- "Node name for S&R": "LTXVImgToVideo"
+ "Node name for S&R": "LTXVImgToVideo",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
768,
512,
97,
+ 1,
1
]
},
@@ -148,11 +167,12 @@
154
],
"flags": {},
- "order": 15,
+ "order": 14,
"mode": 0,
"inputs": [
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"shape": 7,
"link": 185
@@ -161,6 +181,7 @@
"outputs": [
{
"name": "SIGMAS",
+ "localized_name": "SIGMAS",
"type": "SIGMAS",
"links": [
197
@@ -169,7 +190,9 @@
}
],
"properties": {
- "Node name for S&R": "LTXVScheduler"
+ "Node name for S&R": "LTXVScheduler",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
20,
@@ -179,44 +202,6 @@
0.1
]
},
- {
- "id": 79,
- "type": "LTXVApplySTG",
- "pos": [
- 463.5073547363281,
- 87.92158508300781
- ],
- "size": [
- 315,
- 58
- ],
- "flags": {},
- "order": 7,
- "mode": 0,
- "inputs": [
- {
- "name": "model",
- "type": "MODEL",
- "link": 191
- }
- ],
- "outputs": [
- {
- "name": "model",
- "type": "MODEL",
- "links": [
- 195
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "Node name for S&R": "LTXVApplySTG"
- },
- "widgets_values": [
- "14"
- ]
- },
{
"id": 8,
"type": "VAEDecode",
@@ -229,16 +214,18 @@
46
],
"flags": {},
- "order": 18,
+ "order": 17,
"mode": 0,
"inputs": [
{
"name": "samples",
+ "localized_name": "samples",
"type": "LATENT",
"link": 199
},
{
"name": "vae",
+ "localized_name": "vae",
"type": "VAE",
"link": 212
}
@@ -246,6 +233,7 @@
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
211
@@ -254,144 +242,12 @@
}
],
"properties": {
- "Node name for S&R": "VAEDecode"
+ "Node name for S&R": "VAEDecode",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": []
},
- {
- "id": 83,
- "type": "RandomNoise",
- "pos": [
- 464.678955078125,
- -74.81466674804688
- ],
- "size": [
- 315,
- 82
- ],
- "flags": {},
- "order": 0,
- "mode": 0,
- "inputs": [],
- "outputs": [
- {
- "name": "NOISE",
- "type": "NOISE",
- "links": [
- 194
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "Node name for S&R": "RandomNoise"
- },
- "widgets_values": [
- 42,
- "fixed"
- ]
- },
- {
- "id": 69,
- "type": "LTXVConditioning",
- "pos": [
- 885.63818359375,
- 226.95408630371094
- ],
- "size": [
- 223.8660125732422,
- 78
- ],
- "flags": {},
- "order": 14,
- "mode": 0,
- "inputs": [
- {
- "name": "positive",
- "type": "CONDITIONING",
- "link": 183
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "link": 184
- }
- ],
- "outputs": [
- {
- "name": "positive",
- "type": "CONDITIONING",
- "links": [
- 201
- ],
- "slot_index": 0
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "links": [
- 202
- ],
- "slot_index": 1
- }
- ],
- "properties": {
- "Node name for S&R": "LTXVConditioning"
- },
- "widgets_values": [
- 25
- ]
- },
- {
- "id": 82,
- "type": "STGGuider",
- "pos": [
- 830.7182006835938,
- 5.0998334884643555
- ],
- "size": [
- 315,
- 146
- ],
- "flags": {},
- "order": 16,
- "mode": 0,
- "inputs": [
- {
- "name": "model",
- "type": "MODEL",
- "link": 195
- },
- {
- "name": "positive",
- "type": "CONDITIONING",
- "link": 201
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "link": 202
- }
- ],
- "outputs": [
- {
- "name": "GUIDER",
- "type": "GUIDER",
- "links": [
- 200
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "Node name for S&R": "STGGuider"
- },
- "widgets_values": [
- 3,
- 1,
- 0.75
- ]
- },
{
"id": 73,
"type": "KSamplerSelect",
@@ -404,12 +260,13 @@
58
],
"flags": {},
- "order": 1,
+ "order": 0,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "SAMPLER",
+ "localized_name": "SAMPLER",
"type": "SAMPLER",
"links": [
196
@@ -418,7 +275,9 @@
}
],
"properties": {
- "Node name for S&R": "KSamplerSelect"
+ "Node name for S&R": "KSamplerSelect",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
"gradient_estimation"
@@ -436,29 +295,33 @@
520.6666870117188
],
"flags": {},
- "order": 19,
+ "order": 18,
"mode": 0,
"inputs": [
{
"name": "images",
+ "localized_name": "images",
"type": "IMAGE",
"shape": 7,
"link": 211
},
{
"name": "audio",
+ "localized_name": "audio",
"type": "AUDIO",
"shape": 7,
"link": null
},
{
"name": "meta_batch",
+ "localized_name": "meta_batch",
"type": "VHS_BatchManager",
"shape": 7,
"link": null
},
{
"name": "vae",
+ "localized_name": "vae",
"type": "VAE",
"shape": 7,
"link": null
@@ -467,6 +330,7 @@
"outputs": [
{
"name": "Filenames",
+ "localized_name": "Filenames",
"type": "VHS_FILENAMES",
"links": null
}
@@ -488,7 +352,7 @@
"hidden": false,
"paused": false,
"params": {
- "filename": "ltxv_00098.mp4",
+ "filename": "ltxv_00129.mp4",
"subfolder": "",
"type": "output",
"format": "video/h264-mp4",
@@ -497,52 +361,6 @@
}
}
},
- {
- "id": 44,
- "type": "CheckpointLoaderSimple",
- "pos": [
- 82.3931884765625,
- -72.30274200439453
- ],
- "size": [
- 315,
- 98
- ],
- "flags": {},
- "order": 2,
- "mode": 0,
- "inputs": [],
- "outputs": [
- {
- "name": "MODEL",
- "type": "MODEL",
- "links": [
- 191
- ],
- "slot_index": 0
- },
- {
- "name": "CLIP",
- "type": "CLIP",
- "links": null
- },
- {
- "name": "VAE",
- "type": "VAE",
- "links": [
- 212,
- 213
- ],
- "slot_index": 2
- }
- ],
- "properties": {
- "Node name for S&R": "CheckpointLoaderSimple"
- },
- "widgets_values": [
- "ltx-video-2b-v0.9.5.safetensors"
- ]
- },
{
"id": 38,
"type": "CLIPLoader",
@@ -552,15 +370,16 @@
],
"size": [
315,
- 82
+ 106
],
"flags": {},
- "order": 3,
+ "order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "CLIP",
+ "localized_name": "CLIP",
"type": "CLIP",
"links": [
74,
@@ -570,7 +389,9 @@
}
],
"properties": {
- "Node name for S&R": "CLIPLoader"
+ "Node name for S&R": "CLIPLoader",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
"t5xxl_fp16.safetensors",
@@ -590,12 +411,13 @@
82
],
"flags": {},
- "order": 4,
+ "order": 2,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "prompt_enhancer",
+ "localized_name": "prompt_enhancer",
"type": "LTXV_PROMPT_ENHANCER",
"links": [
207
@@ -623,7 +445,7 @@
192.86129760742188
],
"flags": {},
- "order": 5,
+ "order": 3,
"mode": 0,
"inputs": [],
"outputs": [],
@@ -648,17 +470,22 @@
"flags": {
"collapsed": false
},
- "order": 11,
+ "order": 10,
"mode": 0,
"inputs": [
{
"name": "clip",
+ "localized_name": "clip",
"type": "CLIP",
"link": 74
},
{
"name": "text",
"type": "STRING",
+ "pos": [
+ 10,
+ 36
+ ],
"widget": {
"name": "text"
},
@@ -668,6 +495,7 @@
"outputs": [
{
"name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
"type": "CONDITIONING",
"links": [
187
@@ -677,7 +505,9 @@
],
"title": "CLIP Text Encode (Positive Prompt)",
"properties": {
- "Node name for S&R": "CLIPTextEncode"
+ "Node name for S&R": "CLIPTextEncode",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
"best quality, 4k, HDR, a tracking shot of a beautiful scene of the sea waves on the beach with a massive explosion in the water"
@@ -694,19 +524,21 @@
],
"size": [
315,
- 266
+ 286
],
"flags": {},
- "order": 10,
+ "order": 9,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 214
},
{
"name": "get_image_size",
+ "localized_name": "get_image_size",
"type": "IMAGE",
"shape": 7,
"link": null
@@ -715,6 +547,10 @@
"name": "width_input",
"type": "INT",
"shape": 7,
+ "pos": [
+ 10,
+ 196
+ ],
"widget": {
"name": "width_input"
},
@@ -724,6 +560,10 @@
"name": "height_input",
"type": "INT",
"shape": 7,
+ "pos": [
+ 10,
+ 220
+ ],
"widget": {
"name": "height_input"
},
@@ -733,6 +573,7 @@
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
215
@@ -741,11 +582,13 @@
},
{
"name": "width",
+ "localized_name": "width",
"type": "INT",
"links": null
},
{
"name": "height",
+ "localized_name": "height",
"type": "INT",
"links": null
}
@@ -776,11 +619,12 @@
58
],
"flags": {},
- "order": 12,
+ "order": 11,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 215
}
@@ -788,6 +632,7 @@
"outputs": [
{
"name": "output_image",
+ "localized_name": "output_image",
"type": "IMAGE",
"links": [
204
@@ -796,7 +641,9 @@
}
],
"properties": {
- "Node name for S&R": "LTXVPreprocess"
+ "Node name for S&R": "LTXVPreprocess",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
40
@@ -814,11 +661,12 @@
137.85621643066406
],
"flags": {},
- "order": 8,
+ "order": 7,
"mode": 0,
"inputs": [
{
"name": "clip",
+ "localized_name": "clip",
"type": "CLIP",
"link": 75
}
@@ -826,6 +674,7 @@
"outputs": [
{
"name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
"type": "CONDITIONING",
"links": [
188
@@ -835,7 +684,9 @@
],
"title": "CLIP Text Encode (Negative Prompt)",
"properties": {
- "Node name for S&R": "CLIPTextEncode"
+ "Node name for S&R": "CLIPTextEncode",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
"low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly"
@@ -855,12 +706,13 @@
333.3305358886719
],
"flags": {},
- "order": 6,
+ "order": 4,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
205,
@@ -870,12 +722,15 @@
},
{
"name": "MASK",
+ "localized_name": "MASK",
"type": "MASK",
"links": null
}
],
"properties": {
- "Node name for S&R": "LoadImage"
+ "Node name for S&R": "LoadImage",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
"shrek2.jpg",
@@ -894,16 +749,18 @@
102
],
"flags": {},
- "order": 9,
+ "order": 8,
"mode": 0,
"inputs": [
{
"name": "prompt_enhancer",
+ "localized_name": "prompt_enhancer",
"type": "LTXV_PROMPT_ENHANCER",
"link": 207
},
{
"name": "image_prompt",
+ "localized_name": "image_prompt",
"type": "IMAGE",
"shape": 7,
"link": 205
@@ -912,6 +769,7 @@
"outputs": [
{
"name": "str",
+ "localized_name": "str",
"type": "STRING",
"links": [
206
@@ -926,6 +784,208 @@
"a person looking like Shrek talking",
256
]
+ },
+ {
+ "id": 83,
+ "type": "RandomNoise",
+ "pos": [
+ 453.44482421875,
+ 47.46747970581055
+ ],
+ "size": [
+ 315,
+ 82
+ ],
+ "flags": {},
+ "order": 5,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "NOISE",
+ "localized_name": "NOISE",
+ "type": "NOISE",
+ "links": [
+ 194
+ ],
+ "slot_index": 0
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "RandomNoise",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
+ },
+ "widgets_values": [
+ 42,
+ "fixed"
+ ]
+ },
+ {
+ "id": 69,
+ "type": "LTXVConditioning",
+ "pos": [
+ 885.63818359375,
+ 226.95408630371094
+ ],
+ "size": [
+ 223.8660125732422,
+ 78
+ ],
+ "flags": {},
+ "order": 13,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 183
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 184
+ }
+ ],
+ "outputs": [
+ {
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "links": [
+ 221
+ ],
+ "slot_index": 0
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "links": [
+ 222
+ ],
+ "slot_index": 1
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "LTXVConditioning",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
+ },
+ "widgets_values": [
+ 25
+ ]
+ },
+ {
+ "id": 44,
+ "type": "CheckpointLoaderSimple",
+ "pos": [
+ 82.3931884765625,
+ -72.30274200439453
+ ],
+ "size": [
+ 315,
+ 98
+ ],
+ "flags": {},
+ "order": 6,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "MODEL",
+ "localized_name": "MODEL",
+ "type": "MODEL",
+ "links": [
+ 223
+ ],
+ "slot_index": 0
+ },
+ {
+ "name": "CLIP",
+ "localized_name": "CLIP",
+ "type": "CLIP",
+ "links": null
+ },
+ {
+ "name": "VAE",
+ "localized_name": "VAE",
+ "type": "VAE",
+ "links": [
+ 212,
+ 213
+ ],
+ "slot_index": 2
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "CheckpointLoaderSimple",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
+ },
+ "widgets_values": [
+ "ltxv-2b-0.9.6-dev-04-25.safetensors"
+ ]
+ },
+ {
+ "id": 92,
+ "type": "STGGuiderAdvanced",
+ "pos": [
+ 825.4796142578125,
+ -92.4762191772461
+ ],
+ "size": [
+ 327.5999755859375,
+ 242
+ ],
+ "flags": {},
+ "order": 15,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "model",
+ "localized_name": "model",
+ "type": "MODEL",
+ "link": 223
+ },
+ {
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 221
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 222
+ }
+ ],
+ "outputs": [
+ {
+ "name": "GUIDER",
+ "localized_name": "GUIDER",
+ "type": "GUIDER",
+ "links": [
+ 224
+ ],
+ "slot_index": 0
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "STGGuiderAdvanced"
+ },
+ "widgets_values": [
+ 0.996,
+ true,
+ "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "4, 4, 4, 4, 1, 1",
+ "2, 2, 2, 2, 1, 0",
+ "1, 1, 1, 1, 1, 1",
+ "[14], [14], [14], [14], [14], [14]"
+ ]
}
],
"links": [
@@ -985,14 +1045,6 @@
1,
"CONDITIONING"
],
- [
- 191,
- 44,
- 0,
- 79,
- 0,
- "MODEL"
- ],
[
194,
83,
@@ -1001,14 +1053,6 @@
0,
"NOISE"
],
- [
- 195,
- 79,
- 0,
- 82,
- 0,
- "MODEL"
- ],
[
196,
73,
@@ -1041,30 +1085,6 @@
0,
"LATENT"
],
- [
- 200,
- 82,
- 0,
- 81,
- 1,
- "GUIDER"
- ],
- [
- 201,
- 69,
- 0,
- 82,
- 1,
- "CONDITIONING"
- ],
- [
- 202,
- 69,
- 1,
- 82,
- 2,
- "CONDITIONING"
- ],
[
204,
84,
@@ -1136,20 +1156,52 @@
84,
0,
"IMAGE"
+ ],
+ [
+ 221,
+ 69,
+ 0,
+ 92,
+ 1,
+ "CONDITIONING"
+ ],
+ [
+ 222,
+ 69,
+ 1,
+ 92,
+ 2,
+ "CONDITIONING"
+ ],
+ [
+ 223,
+ 44,
+ 0,
+ 92,
+ 0,
+ "MODEL"
+ ],
+ [
+ 224,
+ 92,
+ 0,
+ 81,
+ 1,
+ "GUIDER"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
- "scale": 0.6538676505623292,
+ "scale": 0.6303940863128502,
"offset": [
- 400.85067622291626,
- 9.013459685364836
+ 471.01810112868446,
+ 215.44620489151183
]
},
"node_versions": {
- "comfy-core": "0.3.18"
+ "comfy-core": "0.3.27"
},
"prompt": {
"6": {
@@ -1210,7 +1262,7 @@
},
"44": {
"inputs": {
- "ckpt_name": "ltx-video-2b-v0.9.5.safetensors"
+ "ckpt_name": "ltxv-2b-0.9.6-dev-04-25.safetensors"
},
"class_type": "CheckpointLoaderSimple",
"_meta": {
@@ -1266,6 +1318,7 @@
"height": 512,
"length": 97,
"batch_size": 1,
+ "strength": 1,
"positive": [
"6",
0
@@ -1298,19 +1351,6 @@
"title": "Load Image"
}
},
- "79": {
- "inputs": {
- "block_indices": "14",
- "model": [
- "44",
- 0
- ]
- },
- "class_type": "LTXVApplySTG",
- "_meta": {
- "title": "🅛🅣🅧 LTXV Apply STG"
- }
- },
"81": {
"inputs": {
"noise": [
@@ -1318,7 +1358,7 @@
0
],
"guider": [
- "82",
+ "92",
0
],
"sampler": [
@@ -1339,29 +1379,6 @@
"title": "SamplerCustomAdvanced"
}
},
- "82": {
- "inputs": {
- "cfg": 3,
- "stg": 1,
- "rescale": 0.75,
- "model": [
- "79",
- 0
- ],
- "positive": [
- "69",
- 0
- ],
- "negative": [
- "69",
- 1
- ]
- },
- "class_type": "STGGuider",
- "_meta": {
- "title": "🅛🅣🅧 STG Guider"
- }
- },
"83": {
"inputs": {
"noise_seed": 42
@@ -1450,9 +1467,36 @@
"_meta": {
"title": "Resize Image"
}
+ },
+ "92": {
+ "inputs": {
+ "skip_steps_sigma_threshold": 0.996,
+ "cfg_star_rescale": true,
+ "sigmas": "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "cfg_values": "4, 4, 4, 4, 1, 1",
+ "stg_scale_values": "2, 2, 2, 2, 1, 0",
+ "stg_rescale_values": "1, 1, 1, 1, 1, 1",
+ "stg_layers_indices": "[14], [14], [14], [14], [14], [14]",
+ "model": [
+ "44",
+ 0
+ ],
+ "positive": [
+ "69",
+ 0
+ ],
+ "negative": [
+ "69",
+ 1
+ ]
+ },
+ "class_type": "STGGuiderAdvanced",
+ "_meta": {
+ "title": "🅛🅣🅧 STG Guider Advanced"
+ }
}
},
- "comfy_fork_version": "update_nodes@db0f122d"
+ "comfy_fork_version": "feature/memory_prof@f7e64946"
},
"version": 0.4
}
diff --git a/assets/ltxvideo-i2v.png b/assets/ltxvideo-i2v.png
index 5ce64b0..f2e46d1 100644
--- a/assets/ltxvideo-i2v.png
+++ b/assets/ltxvideo-i2v.png
@@ -1,3 +1,3 @@
version https://git-lfs.github.com/spec/v1
-oid sha256:3e29e3b630ab07c3404e715dd0f5d62ebdb4ef24af4279239e546f2a401824a3
-size 1167784
+oid sha256:5cb7cbf30c7ad19dda71322c40e7a471697350b6ec6b17c8ab14654dad18b909
+size 1142742
diff --git a/assets/ltxvideo-last-sequence-conditioning.json b/assets/ltxvideo-last-sequence-conditioning.json
index ced29ae..4b895b7 100644
--- a/assets/ltxvideo-last-sequence-conditioning.json
+++ b/assets/ltxvideo-last-sequence-conditioning.json
@@ -1,57 +1,7 @@
{
- "last_node_id": 212,
- "last_link_id": 579,
+ "last_node_id": 213,
+ "last_link_id": 584,
"nodes": [
- {
- "id": 191,
- "type": "STGGuider",
- "pos": [
- 1216.2901611328125,
- -1.7041345834732056
- ],
- "size": [
- 389.04827880859375,
- 171.58651733398438
- ],
- "flags": {},
- "order": 19,
- "mode": 0,
- "inputs": [
- {
- "name": "model",
- "type": "MODEL",
- "link": 516
- },
- {
- "name": "positive",
- "type": "CONDITIONING",
- "link": 561
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "link": 562
- }
- ],
- "outputs": [
- {
- "name": "GUIDER",
- "type": "GUIDER",
- "links": [
- 544
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "Node name for S&R": "STGGuider"
- },
- "widgets_values": [
- 3.5,
- 1,
- 0.75
- ]
- },
{
"id": 71,
"type": "LTXVScheduler",
@@ -64,11 +14,12 @@
154
],
"flags": {},
- "order": 20,
+ "order": 19,
"mode": 0,
"inputs": [
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"shape": 7,
"link": 564
@@ -77,6 +28,7 @@
"outputs": [
{
"name": "SIGMAS",
+ "localized_name": "SIGMAS",
"type": "SIGMAS",
"links": [
542
@@ -95,76 +47,6 @@
0.1
]
},
- {
- "id": 188,
- "type": "LTXVApplySTG",
- "pos": [
- 77.9090805053711,
- -126.5878677368164
- ],
- "size": [
- 315,
- 58
- ],
- "flags": {},
- "order": 8,
- "mode": 0,
- "inputs": [
- {
- "name": "model",
- "type": "MODEL",
- "link": 508
- }
- ],
- "outputs": [
- {
- "name": "model",
- "type": "MODEL",
- "links": [
- 516
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "Node name for S&R": "LTXVApplySTG"
- },
- "widgets_values": [
- "14"
- ]
- },
- {
- "id": 73,
- "type": "KSamplerSelect",
- "pos": [
- 471.1808166503906,
- -136.36605834960938
- ],
- "size": [
- 315,
- 58
- ],
- "flags": {},
- "order": 0,
- "mode": 0,
- "inputs": [],
- "outputs": [
- {
- "name": "SAMPLER",
- "type": "SAMPLER",
- "links": [
- 541
- ],
- "slot_index": 0
- }
- ],
- "properties": {
- "Node name for S&R": "KSamplerSelect"
- },
- "widgets_values": [
- "gradient_estimation"
- ]
- },
{
"id": 69,
"type": "LTXVConditioning",
@@ -177,16 +59,18 @@
78
],
"flags": {},
- "order": 17,
+ "order": 16,
"mode": 0,
"inputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"link": 266
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"link": 267
}
@@ -194,6 +78,7 @@
"outputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"links": [
556
@@ -202,6 +87,7 @@
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"links": [
557
@@ -216,52 +102,6 @@
25
]
},
- {
- "id": 44,
- "type": "CheckpointLoaderSimple",
- "pos": [
- -301.724853515625,
- -148.44154357910156
- ],
- "size": [
- 315,
- 98
- ],
- "flags": {},
- "order": 1,
- "mode": 0,
- "inputs": [],
- "outputs": [
- {
- "name": "MODEL",
- "type": "MODEL",
- "links": [
- 508
- ],
- "slot_index": 0
- },
- {
- "name": "CLIP",
- "type": "CLIP",
- "links": null
- },
- {
- "name": "VAE",
- "type": "VAE",
- "links": [
- 565,
- 566
- ],
- "slot_index": 2
- }
- ],
- "properties": {
- "Node name for S&R": "CheckpointLoaderSimple"
- },
- "widgets_values": [
- "ltx-video-2b-v0.9.5.safetensors"
- ]
- },
{
"id": 199,
"type": "LTXVPromptEnhancerLoader",
@@ -274,12 +114,13 @@
82
],
"flags": {},
- "order": 2,
+ "order": 0,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "prompt_enhancer",
+ "localized_name": "prompt_enhancer",
"type": "LTXV_PROMPT_ENHANCER",
"links": [
538
@@ -304,15 +145,16 @@
],
"size": [
315,
- 82
+ 98
],
"flags": {},
- "order": 3,
+ "order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "CLIP",
+ "localized_name": "CLIP",
"type": "CLIP",
"links": [
74,
@@ -344,11 +186,12 @@
"flags": {
"collapsed": true
},
- "order": 14,
+ "order": 13,
"mode": 0,
"inputs": [
{
"name": "clip",
+ "localized_name": "clip",
"type": "CLIP",
"link": 74
},
@@ -364,6 +207,7 @@
"outputs": [
{
"name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
"type": "CONDITIONING",
"links": [
266
@@ -395,11 +239,12 @@
"flags": {
"collapsed": true
},
- "order": 9,
+ "order": 8,
"mode": 0,
"inputs": [
{
"name": "clip",
+ "localized_name": "clip",
"type": "CLIP",
"link": 75
}
@@ -407,6 +252,7 @@
"outputs": [
{
"name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
"type": "CONDITIONING",
"links": [
267
@@ -436,16 +282,18 @@
102
],
"flags": {},
- "order": 10,
+ "order": 9,
"mode": 0,
"inputs": [
{
"name": "prompt_enhancer",
+ "localized_name": "prompt_enhancer",
"type": "LTXV_PROMPT_ENHANCER",
"link": 538
},
{
"name": "image_prompt",
+ "localized_name": "image_prompt",
"type": "IMAGE",
"shape": 7,
"link": 523
@@ -454,6 +302,7 @@
"outputs": [
{
"name": "str",
+ "localized_name": "str",
"type": "STRING",
"links": [
507
@@ -481,11 +330,12 @@
86
],
"flags": {},
- "order": 13,
+ "order": 12,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 567
}
@@ -493,11 +343,13 @@
"outputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"links": null
},
{
- "name": "768 width",
+ "name": "width",
+ "localized_name": "width",
"type": "INT",
"links": [
568
@@ -505,7 +357,8 @@
"slot_index": 1
},
{
- "name": "512 height",
+ "name": "height",
+ "localized_name": "height",
"type": "INT",
"links": [
569
@@ -513,7 +366,8 @@
"slot_index": 2
},
{
- "name": "65 count",
+ "name": "count",
+ "localized_name": "count",
"type": "INT",
"links": [
570
@@ -535,15 +389,19 @@
],
"size": [
315,
- 130
+ 170
],
"flags": {},
- "order": 16,
+ "order": 15,
"mode": 0,
"inputs": [
{
"name": "width",
"type": "INT",
+ "pos": [
+ 10,
+ 36
+ ],
"widget": {
"name": "width"
},
@@ -552,6 +410,10 @@
{
"name": "height",
"type": "INT",
+ "pos": [
+ 10,
+ 60
+ ],
"widget": {
"name": "height"
},
@@ -560,6 +422,10 @@
{
"name": "length",
"type": "INT",
+ "pos": [
+ 10,
+ 84
+ ],
"widget": {
"name": "length"
},
@@ -569,6 +435,7 @@
"outputs": [
{
"name": "LATENT",
+ "localized_name": "LATENT",
"type": "LATENT",
"links": [
558
@@ -598,11 +465,12 @@
58
],
"flags": {},
- "order": 15,
+ "order": 14,
"mode": 0,
"inputs": [
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 571
}
@@ -610,6 +478,7 @@
"outputs": [
{
"name": "output_image",
+ "localized_name": "output_image",
"type": "IMAGE",
"links": [
572
@@ -636,7 +505,7 @@
93.54439544677734
],
"flags": {},
- "order": 4,
+ "order": 2,
"mode": 0,
"inputs": [],
"outputs": [],
@@ -647,79 +516,6 @@
"color": "#432",
"bgcolor": "#653"
},
- {
- "id": 195,
- "type": "VHS_VideoCombine",
- "pos": [
- 2006.0511474609375,
- 61.66352081298828
- ],
- "size": [
- 315,
- 352
- ],
- "flags": {},
- "order": 25,
- "mode": 0,
- "inputs": [
- {
- "name": "images",
- "type": "IMAGE",
- "shape": 7,
- "link": 530
- },
- {
- "name": "audio",
- "type": "AUDIO",
- "shape": 7,
- "link": null
- },
- {
- "name": "meta_batch",
- "type": "VHS_BatchManager",
- "shape": 7,
- "link": null
- },
- {
- "name": "vae",
- "type": "VAE",
- "shape": 7,
- "link": null
- }
- ],
- "outputs": [
- {
- "name": "Filenames",
- "type": "VHS_FILENAMES",
- "links": null
- }
- ],
- "properties": {
- "Node name for S&R": "VHS_VideoCombine"
- },
- "widgets_values": {
- "frame_rate": 25,
- "loop_count": 0,
- "filename_prefix": "ltxv",
- "format": "video/h264-mp4",
- "pix_fmt": "yuv420p",
- "crf": 19,
- "save_metadata": true,
- "pingpong": false,
- "save_output": true,
- "videopreview": {
- "hidden": false,
- "paused": false,
- "params": {
- "filename": "ltxv_00114.mp4",
- "subfolder": "",
- "type": "output",
- "format": "video/h264-mp4",
- "frame_rate": 25
- }
- }
- }
- },
{
"id": 200,
"type": "SamplerCustomAdvanced",
@@ -732,31 +528,36 @@
106
],
"flags": {},
- "order": 21,
+ "order": 20,
"mode": 0,
"inputs": [
{
"name": "noise",
+ "localized_name": "noise",
"type": "NOISE",
"link": 540
},
{
"name": "guider",
+ "localized_name": "guider",
"type": "GUIDER",
- "link": 544
+ "link": 584
},
{
"name": "sampler",
+ "localized_name": "sampler",
"type": "SAMPLER",
"link": 541
},
{
"name": "sigmas",
+ "localized_name": "sigmas",
"type": "SIGMAS",
"link": 542
},
{
"name": "latent_image",
+ "localized_name": "latent_image",
"type": "LATENT",
"link": 563
}
@@ -764,6 +565,7 @@
"outputs": [
{
"name": "output",
+ "localized_name": "output",
"type": "LATENT",
"links": [
578
@@ -772,6 +574,7 @@
},
{
"name": "denoised_output",
+ "localized_name": "denoised_output",
"type": "LATENT",
"links": null
}
@@ -782,222 +585,234 @@
"widgets_values": []
},
{
- "id": 212,
- "type": "LTXVCropGuides",
+ "id": 192,
+ "type": "VHS_SelectImages",
"pos": [
- 1656.453369140625,
- 31.322202682495117
+ 80.78836059570312,
+ 417.3809509277344
],
"size": [
- 216.59999084472656,
- 66
+ 315,
+ 106
],
"flags": {},
- "order": 22,
+ "order": 10,
"mode": 0,
"inputs": [
{
- "name": "positive",
- "type": "CONDITIONING",
- "link": 576
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "link": 577
- },
- {
- "name": "latent",
- "type": "LATENT",
- "link": 578
+ "name": "image",
+ "localized_name": "image",
+ "type": "IMAGE",
+ "link": 525
}
],
"outputs": [
{
- "name": "positive",
- "type": "CONDITIONING",
- "links": null
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "links": null
- },
- {
- "name": "latent",
- "type": "LATENT",
+ "name": "IMAGE",
+ "localized_name": "IMAGE",
+ "type": "IMAGE",
"links": [
- 579
+ 571
],
- "slot_index": 2
+ "slot_index": 0
}
],
"properties": {
- "Node name for S&R": "LTXVCropGuides"
+ "Node name for S&R": "VHS_SelectImages"
},
- "widgets_values": []
+ "widgets_values": {
+ "indexes": ":9",
+ "err_if_missing": true,
+ "err_if_empty": true
+ }
},
{
- "id": 8,
- "type": "VAEDecode",
+ "id": 194,
+ "type": "VHS_SelectImages",
"pos": [
- 1660.500244140625,
- 168.7158966064453
+ 459.2135314941406,
+ 416.92822265625
],
"size": [
- 210,
- 46
+ 315,
+ 106
],
"flags": {},
- "order": 23,
+ "order": 11,
"mode": 0,
"inputs": [
{
- "name": "samples",
- "type": "LATENT",
- "link": 579
- },
- {
- "name": "vae",
- "type": "VAE",
- "link": 565
+ "name": "image",
+ "localized_name": "image",
+ "type": "IMAGE",
+ "link": 527
}
],
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
- 575
+ 574
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "VAEDecode"
+ "Node name for S&R": "VHS_SelectImages"
},
- "widgets_values": []
+ "widgets_values": {
+ "indexes": "9:",
+ "err_if_missing": true,
+ "err_if_empty": true
+ }
},
{
- "id": 193,
- "type": "ImageBatch",
+ "id": 208,
+ "type": "Note",
"pos": [
- 1667.016845703125,
- 302.3862609863281
+ 463.1730651855469,
+ 573.6937255859375
],
"size": [
- 210,
- 46
+ 299.7644958496094,
+ 93.54439544677734
],
"flags": {},
- "order": 24,
+ "order": 3,
"mode": 0,
- "inputs": [
- {
- "name": "image1",
- "type": "IMAGE",
- "link": 575
- },
- {
- "name": "image2",
- "type": "IMAGE",
- "link": 574
- }
- ],
- "outputs": [
- {
- "name": "IMAGE",
- "type": "IMAGE",
- "links": [
- 530
- ],
- "slot_index": 0
- }
+ "inputs": [],
+ "outputs": [],
+ "properties": {},
+ "widgets_values": [
+ "Selects the rest of the video and attaches the generated video to it at the beginning."
],
- "properties": {
- "Node name for S&R": "ImageBatch"
- },
- "widgets_values": []
+ "color": "#432",
+ "bgcolor": "#653"
},
{
- "id": 192,
- "type": "VHS_SelectImages",
+ "id": 134,
+ "type": "VHS_LoadVideo",
"pos": [
- 80.78836059570312,
- 417.3809509277344
+ -256.13458251953125,
+ 277.4459533691406
],
"size": [
- 315,
- 106
+ 239.82791137695312,
+ 422.55194091796875
],
"flags": {},
- "order": 11,
+ "order": 4,
"mode": 0,
"inputs": [
{
- "name": "image",
- "type": "IMAGE",
- "link": 525
+ "name": "meta_batch",
+ "localized_name": "meta_batch",
+ "type": "VHS_BatchManager",
+ "shape": 7,
+ "link": null
+ },
+ {
+ "name": "vae",
+ "localized_name": "vae",
+ "type": "VAE",
+ "shape": 7,
+ "link": null
}
],
"outputs": [
{
"name": "IMAGE",
+ "localized_name": "IMAGE",
"type": "IMAGE",
"links": [
- 571
+ 523,
+ 525,
+ 527,
+ 567
],
"slot_index": 0
+ },
+ {
+ "name": "frame_count",
+ "localized_name": "frame_count",
+ "type": "INT",
+ "links": null
+ },
+ {
+ "name": "audio",
+ "localized_name": "audio",
+ "type": "AUDIO",
+ "links": null
+ },
+ {
+ "name": "video_info",
+ "localized_name": "video_info",
+ "type": "VHS_VIDEOINFO",
+ "links": null
}
],
"properties": {
- "Node name for S&R": "VHS_SelectImages"
+ "Node name for S&R": "VHS_LoadVideo"
},
"widgets_values": {
- "indexes": ":9",
- "err_if_missing": true,
- "err_if_empty": true
+ "video": "jeep.mp4",
+ "force_rate": 0,
+ "force_size": "Disabled",
+ "custom_width": 512,
+ "custom_height": 512,
+ "frame_load_cap": 65,
+ "skip_first_frames": 0,
+ "select_every_nth": 1,
+ "choose video to upload": "image",
+ "videopreview": {
+ "hidden": false,
+ "paused": false,
+ "params": {
+ "frame_load_cap": 65,
+ "skip_first_frames": 0,
+ "force_rate": 0,
+ "filename": "jeep.mp4",
+ "type": "input",
+ "format": "video/mp4",
+ "select_every_nth": 1
+ }
+ }
}
},
{
- "id": 194,
- "type": "VHS_SelectImages",
+ "id": 73,
+ "type": "KSamplerSelect",
"pos": [
- 459.2135314941406,
- 416.92822265625
+ 839.142333984375,
+ 511.1702575683594
],
"size": [
315,
- 106
+ 58
],
"flags": {},
- "order": 12,
+ "order": 5,
"mode": 0,
- "inputs": [
- {
- "name": "image",
- "type": "IMAGE",
- "link": 527
- }
- ],
+ "inputs": [],
"outputs": [
{
- "name": "IMAGE",
- "type": "IMAGE",
+ "name": "SAMPLER",
+ "localized_name": "SAMPLER",
+ "type": "SAMPLER",
"links": [
- 574
+ 541
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "VHS_SelectImages"
+ "Node name for S&R": "KSamplerSelect"
},
- "widgets_values": {
- "indexes": "9:",
- "err_if_missing": true,
- "err_if_empty": true
- }
+ "widgets_values": [
+ "gradient_estimation"
+ ]
},
{
"id": 206,
@@ -1011,31 +826,36 @@
162
],
"flags": {},
- "order": 18,
+ "order": 17,
"mode": 0,
"inputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"link": 556
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"link": 557
},
{
"name": "vae",
+ "localized_name": "vae",
"type": "VAE",
"link": 566
},
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"link": 558
},
{
"name": "image",
+ "localized_name": "image",
"type": "IMAGE",
"link": 572
}
@@ -1043,24 +863,27 @@
"outputs": [
{
"name": "positive",
+ "localized_name": "positive",
"type": "CONDITIONING",
"links": [
- 561,
- 576
+ 576,
+ 581
],
"slot_index": 0
},
{
"name": "negative",
+ "localized_name": "negative",
"type": "CONDITIONING",
"links": [
- 562,
- 577
+ 577,
+ 582
],
"slot_index": 1
},
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"links": [
563,
@@ -1078,84 +901,301 @@
]
},
{
- "id": 190,
- "type": "RandomNoise",
+ "id": 44,
+ "type": "CheckpointLoaderSimple",
"pos": [
- 831.1925048828125,
- -140.77906799316406
+ -301.724853515625,
+ -148.44154357910156
],
"size": [
315,
- 82
+ 98
],
"flags": {},
- "order": 5,
+ "order": 6,
"mode": 0,
"inputs": [],
"outputs": [
{
- "name": "NOISE",
- "type": "NOISE",
+ "name": "MODEL",
+ "localized_name": "MODEL",
+ "type": "MODEL",
"links": [
- 540
+ 583
+ ],
+ "slot_index": 0
+ },
+ {
+ "name": "CLIP",
+ "localized_name": "CLIP",
+ "type": "CLIP",
+ "links": null
+ },
+ {
+ "name": "VAE",
+ "localized_name": "VAE",
+ "type": "VAE",
+ "links": [
+ 565,
+ 566
+ ],
+ "slot_index": 2
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "CheckpointLoaderSimple"
+ },
+ "widgets_values": [
+ "ltxv-2b-0.9.6-dev-04-25.safetensors"
+ ]
+ },
+ {
+ "id": 213,
+ "type": "STGGuiderAdvanced",
+ "pos": [
+ 1207.9185791015625,
+ -48.71029281616211
+ ],
+ "size": [
+ 327.5999755859375,
+ 242
+ ],
+ "flags": {},
+ "order": 18,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "model",
+ "localized_name": "model",
+ "type": "MODEL",
+ "link": 583
+ },
+ {
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 581
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 582
+ }
+ ],
+ "outputs": [
+ {
+ "name": "GUIDER",
+ "localized_name": "GUIDER",
+ "type": "GUIDER",
+ "links": [
+ 584
+ ],
+ "slot_index": 0
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "STGGuiderAdvanced"
+ },
+ "widgets_values": [
+ 0.996,
+ true,
+ "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "4, 4, 4, 4, 1, 1",
+ "2, 2, 2, 2, 1, 0",
+ "1, 1, 1, 1, 1, 1",
+ "[14], [14], [14], [14], [14], [14]"
+ ]
+ },
+ {
+ "id": 212,
+ "type": "LTXVCropGuides",
+ "pos": [
+ 1632.5416259765625,
+ -108.9360580444336
+ ],
+ "size": [
+ 216.59999084472656,
+ 66
+ ],
+ "flags": {},
+ "order": 21,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 576
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 577
+ },
+ {
+ "name": "latent",
+ "localized_name": "latent",
+ "type": "LATENT",
+ "link": 578
+ }
+ ],
+ "outputs": [
+ {
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "links": null
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "links": null
+ },
+ {
+ "name": "latent",
+ "localized_name": "latent",
+ "type": "LATENT",
+ "links": [
+ 579
+ ],
+ "slot_index": 2
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "LTXVCropGuides"
+ },
+ "widgets_values": []
+ },
+ {
+ "id": 8,
+ "type": "VAEDecode",
+ "pos": [
+ 1632.39990234375,
+ 21.121004104614258
+ ],
+ "size": [
+ 210,
+ 46
+ ],
+ "flags": {},
+ "order": 22,
+ "mode": 0,
+ "inputs": [
+ {
+ "name": "samples",
+ "localized_name": "samples",
+ "type": "LATENT",
+ "link": 579
+ },
+ {
+ "name": "vae",
+ "localized_name": "vae",
+ "type": "VAE",
+ "link": 565
+ }
+ ],
+ "outputs": [
+ {
+ "name": "IMAGE",
+ "localized_name": "IMAGE",
+ "type": "IMAGE",
+ "links": [
+ 575
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "RandomNoise"
+ "Node name for S&R": "VAEDecode"
},
- "widgets_values": [
- 43,
- "fixed"
- ]
+ "widgets_values": []
},
{
- "id": 208,
- "type": "Note",
+ "id": 193,
+ "type": "ImageBatch",
"pos": [
- 463.1730651855469,
- 573.6937255859375
+ 1633.234130859375,
+ 133.8223419189453
],
"size": [
- 299.7644958496094,
- 93.54439544677734
+ 210,
+ 46
],
"flags": {},
- "order": 6,
+ "order": 23,
"mode": 0,
- "inputs": [],
- "outputs": [],
- "properties": {},
- "widgets_values": [
- "Selects the rest of the video and attaches the generated video to it at the beginning."
+ "inputs": [
+ {
+ "name": "image1",
+ "localized_name": "image1",
+ "type": "IMAGE",
+ "link": 575
+ },
+ {
+ "name": "image2",
+ "localized_name": "image2",
+ "type": "IMAGE",
+ "link": 574
+ }
],
- "color": "#432",
- "bgcolor": "#653"
+ "outputs": [
+ {
+ "name": "IMAGE",
+ "localized_name": "IMAGE",
+ "type": "IMAGE",
+ "links": [
+ 530
+ ],
+ "slot_index": 0
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "ImageBatch"
+ },
+ "widgets_values": []
},
{
- "id": 134,
- "type": "VHS_LoadVideo",
+ "id": 195,
+ "type": "VHS_VideoCombine",
"pos": [
- -256.13458251953125,
- 277.4459533691406
+ 1635.779052734375,
+ 243.89697265625
],
"size": [
- 239.82791137695312,
- 422.55194091796875
+ 315,
+ 520.6666870117188
],
"flags": {},
- "order": 7,
+ "order": 24,
"mode": 0,
"inputs": [
+ {
+ "name": "images",
+ "localized_name": "images",
+ "type": "IMAGE",
+ "shape": 7,
+ "link": 530
+ },
+ {
+ "name": "audio",
+ "localized_name": "audio",
+ "type": "AUDIO",
+ "shape": 7,
+ "link": null
+ },
{
"name": "meta_batch",
+ "localized_name": "meta_batch",
"type": "VHS_BatchManager",
"shape": 7,
"link": null
},
{
"name": "vae",
+ "localized_name": "vae",
"type": "VAE",
"shape": 7,
"link": null
@@ -1163,59 +1203,71 @@
],
"outputs": [
{
- "name": "IMAGE",
- "type": "IMAGE",
- "links": [
- 523,
- 525,
- 527,
- 567
- ],
- "slot_index": 0
- },
- {
- "name": "frame_count",
- "type": "INT",
- "links": null
- },
- {
- "name": "audio",
- "type": "AUDIO",
- "links": null
- },
- {
- "name": "video_info",
- "type": "VHS_VIDEOINFO",
+ "name": "Filenames",
+ "localized_name": "Filenames",
+ "type": "VHS_FILENAMES",
"links": null
}
],
"properties": {
- "Node name for S&R": "VHS_LoadVideo"
+ "Node name for S&R": "VHS_VideoCombine"
},
"widgets_values": {
- "video": "jeep.mp4",
- "force_rate": 0,
- "force_size": "Disabled",
- "custom_width": 512,
- "custom_height": 512,
- "frame_load_cap": 65,
- "skip_first_frames": 0,
- "select_every_nth": 1,
- "choose video to upload": "image",
+ "frame_rate": 25,
+ "loop_count": 0,
+ "filename_prefix": "ltxv",
+ "format": "video/h264-mp4",
+ "pix_fmt": "yuv420p",
+ "crf": 19,
+ "save_metadata": true,
+ "pingpong": false,
+ "save_output": true,
"videopreview": {
"hidden": false,
"paused": false,
"params": {
- "frame_load_cap": 65,
- "skip_first_frames": 0,
- "force_rate": 0,
- "filename": "jeep.mp4",
- "type": "input",
- "format": "video/mp4",
- "select_every_nth": 1
+ "filename": "ltxv_00124.mp4",
+ "subfolder": "",
+ "type": "output",
+ "format": "video/h264-mp4",
+ "frame_rate": 25
}
}
}
+ },
+ {
+ "id": 190,
+ "type": "RandomNoise",
+ "pos": [
+ 831.1925048828125,
+ -140.77906799316406
+ ],
+ "size": [
+ 315,
+ 82
+ ],
+ "flags": {},
+ "order": 7,
+ "mode": 0,
+ "inputs": [],
+ "outputs": [
+ {
+ "name": "NOISE",
+ "localized_name": "NOISE",
+ "type": "NOISE",
+ "links": [
+ 540
+ ],
+ "slot_index": 0
+ }
+ ],
+ "properties": {
+ "Node name for S&R": "RandomNoise"
+ },
+ "widgets_values": [
+ 42,
+ "fixed"
+ ]
}
],
"links": [
@@ -1259,22 +1311,6 @@
1,
"STRING"
],
- [
- 508,
- 44,
- 0,
- 188,
- 0,
- "MODEL"
- ],
- [
- 516,
- 188,
- 0,
- 191,
- 0,
- "MODEL"
- ],
[
523,
134,
@@ -1339,14 +1375,6 @@
3,
"SIGMAS"
],
- [
- 544,
- 191,
- 0,
- 200,
- 1,
- "GUIDER"
- ],
[
556,
69,
@@ -1371,22 +1399,6 @@
3,
"LATENT"
],
- [
- 561,
- 206,
- 0,
- 191,
- 1,
- "CONDITIONING"
- ],
- [
- 562,
- 206,
- 1,
- 191,
- 2,
- "CONDITIONING"
- ],
[
563,
206,
@@ -1514,16 +1526,48 @@
8,
0,
"LATENT"
+ ],
+ [
+ 581,
+ 206,
+ 0,
+ 213,
+ 1,
+ "CONDITIONING"
+ ],
+ [
+ 582,
+ 206,
+ 1,
+ 213,
+ 2,
+ "CONDITIONING"
+ ],
+ [
+ 583,
+ 44,
+ 0,
+ 213,
+ 0,
+ "MODEL"
+ ],
+ [
+ 584,
+ 213,
+ 0,
+ 200,
+ 1,
+ "GUIDER"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
- "scale": 0.7117163576202091,
+ "scale": 0.6303940863128502,
"offset": [
- 500.3176701951572,
- 447.6564128377701
+ 544.9425929851524,
+ 236.16737008581478
]
},
"prompt": {
@@ -1585,7 +1629,7 @@
},
"44": {
"inputs": {
- "ckpt_name": "ltx-video-2b-v0.9.5.safetensors"
+ "ckpt_name": "ltxv-2b-0.9.6-dev-04-25.safetensors"
},
"class_type": "CheckpointLoaderSimple",
"_meta": {
@@ -1690,51 +1734,15 @@
"title": "🅛🅣🅧 LTXV Prompt Enhancer"
}
},
- "188": {
- "inputs": {
- "block_indices": "14",
- "model": [
- "44",
- 0
- ]
- },
- "class_type": "LTXVApplySTG",
- "_meta": {
- "title": "🅛🅣🅧 LTXV Apply STG"
- }
- },
"190": {
"inputs": {
- "noise_seed": 43
+ "noise_seed": 42
},
"class_type": "RandomNoise",
"_meta": {
"title": "RandomNoise"
}
},
- "191": {
- "inputs": {
- "cfg": 3.5,
- "stg": 1,
- "rescale": 0.75,
- "model": [
- "188",
- 0
- ],
- "positive": [
- "206",
- 0
- ],
- "negative": [
- "206",
- 1
- ]
- },
- "class_type": "STGGuider",
- "_meta": {
- "title": "🅛🅣🅧 STG Guider"
- }
- },
"192": {
"inputs": {
"indexes": ":9",
@@ -1819,7 +1827,7 @@
0
],
"guider": [
- "191",
+ "213",
0
],
"sampler": [
@@ -1914,11 +1922,38 @@
"_meta": {
"title": "LTXVCropGuides"
}
+ },
+ "213": {
+ "inputs": {
+ "skip_steps_sigma_threshold": 0.996,
+ "cfg_star_rescale": true,
+ "sigmas": "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "cfg_values": "4, 4, 4, 4, 1, 1",
+ "stg_scale_values": "2, 2, 2, 2, 1, 0",
+ "stg_rescale_values": "1, 1, 1, 1, 1, 1",
+ "stg_layers_indices": "[14], [14], [14], [14], [14], [14]",
+ "model": [
+ "44",
+ 0
+ ],
+ "positive": [
+ "206",
+ 0
+ ],
+ "negative": [
+ "206",
+ 1
+ ]
+ },
+ "class_type": "STGGuiderAdvanced",
+ "_meta": {
+ "title": "🅛🅣🅧 STG Guider Advanced"
+ }
}
},
- "comfy_fork_version": "HEAD@e7bd1751",
+ "comfy_fork_version": "feature/memory_prof@f7e64946",
"node_versions": {
- "comfy-core": "0.3.18"
+ "comfy-core": "0.3.27"
}
},
"version": 0.4
diff --git a/assets/ltxvideo-last-sequence-conditioning.png b/assets/ltxvideo-last-sequence-conditioning.png
index 68fb24b..a5b92b9 100644
--- a/assets/ltxvideo-last-sequence-conditioning.png
+++ b/assets/ltxvideo-last-sequence-conditioning.png
@@ -1,3 +1,3 @@
version https://git-lfs.github.com/spec/v1
-oid sha256:7ba0a0ab34867f35268823c6ef8ca4cd8067cd6484e0c6f5b1277c9778bb243e
-size 1379716
+oid sha256:006feae06d8679de9175eae516bf0919b273745859bb045b3d229329a34084fb
+size 1160850
diff --git a/assets/ltxvideo-t2v.json b/assets/ltxvideo-t2v.json
index bc48def..6658ecf 100644
--- a/assets/ltxvideo-t2v.json
+++ b/assets/ltxvideo-t2v.json
@@ -1,119 +1,7 @@
{
- "last_node_id": 88,
- "last_link_id": 212,
+ "last_node_id": 90,
+ "last_link_id": 220,
"nodes": [
- {
- "id": 81,
- "type": "SamplerCustomAdvanced",
- "pos": [
- 1199.310302734375,
- 17.57975959777832
- ],
- "size": [
- 355.20001220703125,
- 106
- ],
- "flags": {},
- "order": 13,
- "mode": 0,
- "inputs": [
- {
- "name": "noise",
- "type": "NOISE",
- "link": 194
- },
- {
- "name": "guider",
- "type": "GUIDER",
- "link": 200
- },
- {
- "name": "sampler",
- "type": "SAMPLER",
- "link": 196
- },
- {
- "name": "sigmas",
- "type": "SIGMAS",
- "link": 197
- },
- {
- "name": "latent_image",
- "type": "LATENT",
- "link": 205
- }
- ],
- "outputs": [
- {
- "name": "output",
- "type": "LATENT",
- "links": [
- 199
- ],
- "slot_index": 0
- },
- {
- "name": "denoised_output",
- "type": "LATENT",
- "links": null
- }
- ],
- "properties": {
- "Node name for S&R": "SamplerCustomAdvanced"
- },
- "widgets_values": []
- },
- {
- "id": 69,
- "type": "LTXVConditioning",
- "pos": [
- 900.875,
- 273.27288818359375
- ],
- "size": [
- 223.8660125732422,
- 78
- ],
- "flags": {},
- "order": 11,
- "mode": 0,
- "inputs": [
- {
- "name": "positive",
- "type": "CONDITIONING",
- "link": 203
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "link": 204
- }
- ],
- "outputs": [
- {
- "name": "positive",
- "type": "CONDITIONING",
- "links": [
- 201
- ],
- "slot_index": 0
- },
- {
- "name": "negative",
- "type": "CONDITIONING",
- "links": [
- 202
- ],
- "slot_index": 1
- }
- ],
- "properties": {
- "Node name for S&R": "LTXVConditioning"
- },
- "widgets_values": [
- 25
- ]
- },
{
"id": 6,
"type": "CLIPTextEncode",
@@ -123,16 +11,17 @@
],
"size": [
295.2641296386719,
- 54
+ 78
],
"flags": {
"collapsed": true
},
- "order": 10,
+ "order": 9,
"mode": 0,
"inputs": [
{
"name": "clip",
+ "localized_name": "clip",
"type": "CLIP",
"link": 74
},
@@ -148,6 +37,7 @@
"outputs": [
{
"name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
"type": "CONDITIONING",
"links": [
203
@@ -157,7 +47,9 @@
],
"title": "CLIP Text Encode (Positive Prompt)",
"properties": {
- "Node name for S&R": "CLIPTextEncode"
+ "Node name for S&R": "CLIPTextEncode",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
""
@@ -179,11 +71,12 @@
"flags": {
"collapsed": true
},
- "order": 9,
+ "order": 8,
"mode": 0,
"inputs": [
{
"name": "clip",
+ "localized_name": "clip",
"type": "CLIP",
"link": 75
}
@@ -191,6 +84,7 @@
"outputs": [
{
"name": "CONDITIONING",
+ "localized_name": "CONDITIONING",
"type": "CONDITIONING",
"links": [
204
@@ -200,7 +94,9 @@
],
"title": "CLIP Text Encode (Negative Prompt)",
"properties": {
- "Node name for S&R": "CLIPTextEncode"
+ "Node name for S&R": "CLIPTextEncode",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
"low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly"
@@ -226,6 +122,7 @@
"outputs": [
{
"name": "prompt_enhancer",
+ "localized_name": "prompt_enhancer",
"type": "LTXV_PROMPT_ENHANCER",
"links": [
206
@@ -242,167 +139,229 @@
]
},
{
- "id": 82,
- "type": "STGGuider",
+ "id": 84,
+ "type": "EmptyLTXVLatentVideo",
"pos": [
- 844.2092895507812,
- 72.4302978515625
+ 865.0157470703125,
+ 427.244384765625
],
"size": [
315,
- 146
+ 130
],
"flags": {},
- "order": 12,
+ "order": 1,
"mode": 0,
- "inputs": [
- {
- "name": "model",
- "type": "MODEL",
- "link": 195
- },
- {
- "name": "positive",
- "type": "CONDITIONING",
- "link": 201
- },
+ "inputs": [],
+ "outputs": [
{
- "name": "negative",
- "type": "CONDITIONING",
- "link": 202
+ "name": "LATENT",
+ "localized_name": "LATENT",
+ "type": "LATENT",
+ "links": [
+ 205,
+ 210
+ ],
+ "slot_index": 0
}
],
+ "properties": {
+ "Node name for S&R": "EmptyLTXVLatentVideo",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
+ },
+ "widgets_values": [
+ 768,
+ 512,
+ 97,
+ 1
+ ]
+ },
+ {
+ "id": 38,
+ "type": "CLIPLoader",
+ "pos": [
+ 73.48588562011719,
+ 131.20065307617188
+ ],
+ "size": [
+ 315,
+ 106
+ ],
+ "flags": {},
+ "order": 2,
+ "mode": 0,
+ "inputs": [],
"outputs": [
{
- "name": "GUIDER",
- "type": "GUIDER",
+ "name": "CLIP",
+ "localized_name": "CLIP",
+ "type": "CLIP",
"links": [
- 200
+ 74,
+ 75
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "STGGuider"
+ "Node name for S&R": "CLIPLoader",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
- 3,
- 1,
- 0.75
+ "t5xxl_fp16.safetensors",
+ "ltxv",
+ "default"
]
},
{
- "id": 73,
- "type": "KSamplerSelect",
+ "id": 83,
+ "type": "RandomNoise",
"pos": [
- 853.5048828125,
- -61.43290328979492
+ 451.1808166503906,
+ -72.91265106201172
],
"size": [
315,
- 58
+ 82
],
"flags": {},
- "order": 1,
+ "order": 3,
"mode": 0,
"inputs": [],
"outputs": [
{
- "name": "SAMPLER",
- "type": "SAMPLER",
+ "name": "NOISE",
+ "localized_name": "NOISE",
+ "type": "NOISE",
"links": [
- 196
+ 194
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "KSamplerSelect"
+ "Node name for S&R": "RandomNoise",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
- "gradient_estimation"
+ 42,
+ "fixed"
]
},
{
- "id": 84,
- "type": "EmptyLTXVLatentVideo",
+ "id": 73,
+ "type": "KSamplerSelect",
"pos": [
- 865.0157470703125,
- 427.244384765625
+ 830.7018432617188,
+ -151.5533905029297
],
"size": [
315,
- 130
+ 58
],
"flags": {},
- "order": 2,
+ "order": 4,
"mode": 0,
"inputs": [],
"outputs": [
{
- "name": "LATENT",
- "type": "LATENT",
+ "name": "SAMPLER",
+ "localized_name": "SAMPLER",
+ "type": "SAMPLER",
"links": [
- 205,
- 210
+ 196
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "EmptyLTXVLatentVideo"
+ "Node name for S&R": "KSamplerSelect",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
- 768,
- 512,
- 97,
- 1
+ "gradient_estimation"
]
},
{
- "id": 79,
- "type": "LTXVApplySTG",
+ "id": 81,
+ "type": "SamplerCustomAdvanced",
"pos": [
- 456.1823425292969,
- 77.06049346923828
+ 1206.2567138671875,
+ 111.3145523071289
],
"size": [
- 315,
- 58
+ 355.20001220703125,
+ 106
],
"flags": {},
- "order": 8,
+ "order": 12,
"mode": 0,
"inputs": [
{
- "name": "model",
- "type": "MODEL",
- "link": 191
+ "name": "noise",
+ "localized_name": "noise",
+ "type": "NOISE",
+ "link": 194
+ },
+ {
+ "name": "guider",
+ "localized_name": "guider",
+ "type": "GUIDER",
+ "link": 218
+ },
+ {
+ "name": "sampler",
+ "localized_name": "sampler",
+ "type": "SAMPLER",
+ "link": 196
+ },
+ {
+ "name": "sigmas",
+ "localized_name": "sigmas",
+ "type": "SIGMAS",
+ "link": 197
+ },
+ {
+ "name": "latent_image",
+ "localized_name": "latent_image",
+ "type": "LATENT",
+ "link": 205
}
],
"outputs": [
{
- "name": "model",
- "type": "MODEL",
+ "name": "output",
+ "localized_name": "output",
+ "type": "LATENT",
"links": [
- 195
+ 199
],
"slot_index": 0
+ },
+ {
+ "name": "denoised_output",
+ "localized_name": "denoised_output",
+ "type": "LATENT",
+ "links": null
}
],
"properties": {
- "Node name for S&R": "LTXVApplySTG"
+ "Node name for S&R": "SamplerCustomAdvanced",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
- "widgets_values": [
- "14"
- ]
+ "widgets_values": []
},
{
"id": 71,
"type": "LTXVScheduler",
"pos": [
- 1215.2900390625,
- 217.12747192382812
+ 1210.519287109375,
+ -119.96339416503906
],
"size": [
315,
@@ -414,6 +373,7 @@
"inputs": [
{
"name": "latent",
+ "localized_name": "latent",
"type": "LATENT",
"shape": 7,
"link": 210
@@ -422,6 +382,7 @@
"outputs": [
{
"name": "SIGMAS",
+ "localized_name": "SIGMAS",
"type": "SIGMAS",
"links": [
197
@@ -430,7 +391,9 @@
}
],
"properties": {
- "Node name for S&R": "LTXVScheduler"
+ "Node name for S&R": "LTXVScheduler",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
20,
@@ -441,276 +404,341 @@
]
},
{
- "id": 86,
- "type": "LTXVPromptEnhancer",
+ "id": 8,
+ "type": "VAEDecode",
"pos": [
- 447.4213562011719,
- 346.47674560546875
+ 1268.183349609375,
+ 275.46270751953125
],
"size": [
- 335.70489501953125,
- 124.35137939453125
+ 210,
+ 46
],
"flags": {},
- "order": 6,
+ "order": 13,
"mode": 0,
"inputs": [
{
- "name": "prompt_enhancer",
- "type": "LTXV_PROMPT_ENHANCER",
- "link": 206
+ "name": "samples",
+ "localized_name": "samples",
+ "type": "LATENT",
+ "link": 199
},
{
- "name": "image_prompt",
- "type": "IMAGE",
- "shape": 7,
- "link": null
+ "name": "vae",
+ "localized_name": "vae",
+ "type": "VAE",
+ "link": 211
}
],
"outputs": [
{
- "name": "str",
- "type": "STRING",
+ "name": "IMAGE",
+ "localized_name": "IMAGE",
+ "type": "IMAGE",
"links": [
- 207
+ 212
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "LTXVPromptEnhancer"
+ "Node name for S&R": "VAEDecode",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
- "widgets_values": [
- "dog",
- 256
- ]
+ "widgets_values": []
},
{
- "id": 44,
- "type": "CheckpointLoaderSimple",
+ "id": 88,
+ "type": "VHS_VideoCombine",
"pos": [
- 93.61398315429688,
- -66.24134063720703
+ 1592.291748046875,
+ -89.84502410888672
],
"size": [
315,
- 98
+ 520.6666870117188
],
"flags": {},
- "order": 3,
+ "order": 14,
"mode": 0,
- "inputs": [],
- "outputs": [
+ "inputs": [
{
- "name": "MODEL",
- "type": "MODEL",
- "links": [
- 191
- ],
- "slot_index": 0
+ "name": "images",
+ "localized_name": "images",
+ "type": "IMAGE",
+ "shape": 7,
+ "link": 212
},
{
- "name": "CLIP",
- "type": "CLIP",
- "links": null
+ "name": "audio",
+ "localized_name": "audio",
+ "type": "AUDIO",
+ "shape": 7,
+ "link": null
},
{
- "name": "VAE",
+ "name": "meta_batch",
+ "localized_name": "meta_batch",
+ "type": "VHS_BatchManager",
+ "shape": 7,
+ "link": null
+ },
+ {
+ "name": "vae",
+ "localized_name": "vae",
"type": "VAE",
- "links": [
- 211
- ],
- "slot_index": 2
+ "shape": 7,
+ "link": null
+ }
+ ],
+ "outputs": [
+ {
+ "name": "Filenames",
+ "localized_name": "Filenames",
+ "type": "VHS_FILENAMES",
+ "links": null
}
],
"properties": {
- "Node name for S&R": "CheckpointLoaderSimple"
+ "Node name for S&R": "VHS_VideoCombine"
},
- "widgets_values": [
- "ltx-video-2b-v0.9.5.safetensors"
- ]
+ "widgets_values": {
+ "frame_rate": 25,
+ "loop_count": 0,
+ "filename_prefix": "ltxv",
+ "format": "video/h264-mp4",
+ "pix_fmt": "yuv420p",
+ "crf": 19,
+ "save_metadata": true,
+ "pingpong": false,
+ "save_output": true,
+ "videopreview": {
+ "hidden": false,
+ "paused": false,
+ "params": {
+ "filename": "ltxv_00128.mp4",
+ "subfolder": "",
+ "type": "output",
+ "format": "video/h264-mp4",
+ "frame_rate": 25
+ }
+ }
+ }
},
{
- "id": 38,
- "type": "CLIPLoader",
+ "id": 44,
+ "type": "CheckpointLoaderSimple",
"pos": [
- 73.48588562011719,
- 131.20065307617188
+ 93.61398315429688,
+ -66.24134063720703
],
"size": [
315,
- 82
+ 98
],
"flags": {},
- "order": 4,
+ "order": 5,
"mode": 0,
"inputs": [],
"outputs": [
+ {
+ "name": "MODEL",
+ "localized_name": "MODEL",
+ "type": "MODEL",
+ "links": [
+ 217
+ ],
+ "slot_index": 0
+ },
{
"name": "CLIP",
+ "localized_name": "CLIP",
"type": "CLIP",
+ "links": null
+ },
+ {
+ "name": "VAE",
+ "localized_name": "VAE",
+ "type": "VAE",
"links": [
- 74,
- 75
+ 211
],
- "slot_index": 0
+ "slot_index": 2
}
],
"properties": {
- "Node name for S&R": "CLIPLoader"
+ "Node name for S&R": "CheckpointLoaderSimple",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
"widgets_values": [
- "t5xxl_fp16.safetensors",
- "ltxv",
- "default"
+ "ltxv-2b-0.9.6-dev-04-25.safetensors"
]
},
{
- "id": 83,
- "type": "RandomNoise",
+ "id": 90,
+ "type": "STGGuiderAdvanced",
"pos": [
- 451.1808166503906,
- -72.91265106201172
+ 830.8892822265625,
+ -37.005123138427734
],
"size": [
- 315,
- 82
+ 327.5999755859375,
+ 242
],
"flags": {},
- "order": 5,
+ "order": 11,
"mode": 0,
- "inputs": [],
+ "inputs": [
+ {
+ "name": "model",
+ "localized_name": "model",
+ "type": "MODEL",
+ "link": 217
+ },
+ {
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 219
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 220
+ }
+ ],
"outputs": [
{
- "name": "NOISE",
- "type": "NOISE",
+ "name": "GUIDER",
+ "localized_name": "GUIDER",
+ "type": "GUIDER",
"links": [
- 194
+ 218
],
"slot_index": 0
}
],
"properties": {
- "Node name for S&R": "RandomNoise"
+ "Node name for S&R": "STGGuiderAdvanced"
},
"widgets_values": [
- 42,
- "fixed"
+ 0.996,
+ true,
+ "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "4, 4, 4, 4, 1, 1",
+ "2, 2, 2, 2, 1, 0",
+ "1, 1, 1, 1, 1, 1",
+ "[14], [14], [14], [14], [14], [14]"
]
},
{
- "id": 8,
- "type": "VAEDecode",
+ "id": 69,
+ "type": "LTXVConditioning",
"pos": [
- 1600,
- 30
+ 900.875,
+ 273.27288818359375
],
"size": [
- 210,
- 46
+ 223.8660125732422,
+ 78
],
"flags": {},
- "order": 14,
+ "order": 10,
"mode": 0,
"inputs": [
{
- "name": "samples",
- "type": "LATENT",
- "link": 199
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
+ "link": 203
},
{
- "name": "vae",
- "type": "VAE",
- "link": 211
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "link": 204
}
],
"outputs": [
{
- "name": "IMAGE",
- "type": "IMAGE",
+ "name": "positive",
+ "localized_name": "positive",
+ "type": "CONDITIONING",
"links": [
- 212
+ 219
],
"slot_index": 0
+ },
+ {
+ "name": "negative",
+ "localized_name": "negative",
+ "type": "CONDITIONING",
+ "links": [
+ 220
+ ],
+ "slot_index": 1
}
],
"properties": {
- "Node name for S&R": "VAEDecode"
+ "Node name for S&R": "LTXVConditioning",
+ "cnr_id": "comfy-core",
+ "ver": "0.3.28"
},
- "widgets_values": []
+ "widgets_values": [
+ 25
+ ]
},
{
- "id": 88,
- "type": "VHS_VideoCombine",
+ "id": 86,
+ "type": "LTXVPromptEnhancer",
"pos": [
- 1862.038330078125,
- 29.338851928710938
+ 447.4213562011719,
+ 346.47674560546875
],
"size": [
- 315,
- 520.6666870117188
+ 335.70489501953125,
+ 124.35137939453125
],
"flags": {},
- "order": 15,
+ "order": 6,
"mode": 0,
"inputs": [
{
- "name": "images",
- "type": "IMAGE",
- "shape": 7,
- "link": 212
- },
- {
- "name": "audio",
- "type": "AUDIO",
- "shape": 7,
- "link": null
- },
- {
- "name": "meta_batch",
- "type": "VHS_BatchManager",
- "shape": 7,
- "link": null
+ "name": "prompt_enhancer",
+ "localized_name": "prompt_enhancer",
+ "type": "LTXV_PROMPT_ENHANCER",
+ "link": 206
},
{
- "name": "vae",
- "type": "VAE",
+ "name": "image_prompt",
+ "localized_name": "image_prompt",
+ "type": "IMAGE",
"shape": 7,
"link": null
}
],
"outputs": [
{
- "name": "Filenames",
- "type": "VHS_FILENAMES",
- "links": null
+ "name": "str",
+ "localized_name": "str",
+ "type": "STRING",
+ "links": [
+ 207
+ ],
+ "slot_index": 0
}
],
"properties": {
- "Node name for S&R": "VHS_VideoCombine"
+ "Node name for S&R": "LTXVPromptEnhancer"
},
- "widgets_values": {
- "frame_rate": 25,
- "loop_count": 0,
- "filename_prefix": "ltxv",
- "format": "video/h264-mp4",
- "pix_fmt": "yuv420p",
- "crf": 19,
- "save_metadata": true,
- "pingpong": false,
- "save_output": true,
- "videopreview": {
- "hidden": false,
- "paused": false,
- "params": {
- "filename": "ltxv_00088.mp4",
- "subfolder": "",
- "type": "output",
- "format": "video/h264-mp4",
- "frame_rate": 25
- }
- }
- }
+ "widgets_values": [
+ "sleeping cat",
+ 256
+ ]
}
],
"links": [
@@ -730,14 +758,6 @@
0,
"CLIP"
],
- [
- 191,
- 44,
- 0,
- 79,
- 0,
- "MODEL"
- ],
[
194,
83,
@@ -746,14 +766,6 @@
0,
"NOISE"
],
- [
- 195,
- 79,
- 0,
- 82,
- 0,
- "MODEL"
- ],
[
196,
73,
@@ -778,30 +790,6 @@
0,
"LATENT"
],
- [
- 200,
- 82,
- 0,
- 81,
- 1,
- "GUIDER"
- ],
- [
- 201,
- 69,
- 0,
- 82,
- 1,
- "CONDITIONING"
- ],
- [
- 202,
- 69,
- 1,
- 82,
- 2,
- "CONDITIONING"
- ],
[
203,
6,
@@ -865,20 +853,52 @@
88,
0,
"IMAGE"
+ ],
+ [
+ 217,
+ 44,
+ 0,
+ 90,
+ 0,
+ "MODEL"
+ ],
+ [
+ 218,
+ 90,
+ 0,
+ 81,
+ 1,
+ "GUIDER"
+ ],
+ [
+ 219,
+ 69,
+ 0,
+ 90,
+ 1,
+ "CONDITIONING"
+ ],
+ [
+ 220,
+ 69,
+ 1,
+ 90,
+ 2,
+ "CONDITIONING"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
- "scale": 0.7865676463226887,
+ "scale": 0.6934334949441353,
"offset": [
- 105.43466179233377,
- 168.14196262252545
+ 217.20361259812148,
+ 342.27826789342566
]
},
"node_versions": {
- "comfy-core": "0.3.18"
+ "comfy-core": "0.3.27"
},
"prompt": {
"6": {
@@ -939,7 +959,7 @@
},
"44": {
"inputs": {
- "ckpt_name": "ltx-video-2b-v0.9.5.safetensors"
+ "ckpt_name": "ltxv-2b-0.9.6-dev-04-25.safetensors"
},
"class_type": "CheckpointLoaderSimple",
"_meta": {
@@ -989,19 +1009,6 @@
"title": "KSamplerSelect"
}
},
- "79": {
- "inputs": {
- "block_indices": "14",
- "model": [
- "44",
- 0
- ]
- },
- "class_type": "LTXVApplySTG",
- "_meta": {
- "title": "🅛🅣🅧 LTXV Apply STG"
- }
- },
"81": {
"inputs": {
"noise": [
@@ -1009,7 +1016,7 @@
0
],
"guider": [
- "82",
+ "90",
0
],
"sampler": [
@@ -1030,29 +1037,6 @@
"title": "SamplerCustomAdvanced"
}
},
- "82": {
- "inputs": {
- "cfg": 3,
- "stg": 1,
- "rescale": 0.75,
- "model": [
- "79",
- 0
- ],
- "positive": [
- "69",
- 0
- ],
- "negative": [
- "69",
- 1
- ]
- },
- "class_type": "STGGuider",
- "_meta": {
- "title": "🅛🅣🅧 STG Guider"
- }
- },
"83": {
"inputs": {
"noise_seed": 42
@@ -1086,7 +1070,7 @@
},
"86": {
"inputs": {
- "prompt": "dog",
+ "prompt": "sleeping cat",
"max_resulting_tokens": 256,
"prompt_enhancer": [
"85",
@@ -1118,9 +1102,36 @@
"_meta": {
"title": "Video Combine 🎥🅥🅗🅢"
}
+ },
+ "90": {
+ "inputs": {
+ "skip_steps_sigma_threshold": 0.996,
+ "cfg_star_rescale": true,
+ "sigmas": "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "cfg_values": "4, 4, 4, 4, 1, 1",
+ "stg_scale_values": "2, 2, 2, 2, 1, 0",
+ "stg_rescale_values": "1, 1, 1, 1, 1, 1",
+ "stg_layers_indices": "[14], [14], [14], [14], [14], [14]",
+ "model": [
+ "44",
+ 0
+ ],
+ "positive": [
+ "69",
+ 0
+ ],
+ "negative": [
+ "69",
+ 1
+ ]
+ },
+ "class_type": "STGGuiderAdvanced",
+ "_meta": {
+ "title": "🅛🅣🅧 STG Guider Advanced"
+ }
}
},
- "comfy_fork_version": "update_nodes@db0f122d"
+ "comfy_fork_version": "feature/memory_prof@f7e64946"
},
"version": 0.4
}
diff --git a/assets/ltxvideo-t2v.png b/assets/ltxvideo-t2v.png
index f7a55b3..8fad809 100644
--- a/assets/ltxvideo-t2v.png
+++ b/assets/ltxvideo-t2v.png
@@ -1,3 +1,3 @@
version https://git-lfs.github.com/spec/v1
-oid sha256:0823a77dec3324a8f696072a3988fe50431ce89482698e09047d32852f886cdc
-size 770420
+oid sha256:7a4f2fe897fd955633099a429aa73bccc2e6007627a3e1bf419b4b8db67bce80
+size 817453
diff --git a/latents.py b/latents.py
new file mode 100644
index 0000000..5323c20
--- /dev/null
+++ b/latents.py
@@ -0,0 +1,351 @@
+from typing import Optional
+
+import comfy_extras.nodes_lt as nodes_lt
+import torch
+
+from .nodes_registry import comfy_node
+
+
+@comfy_node(name="LTXVSelectLatents")
+class LTXVSelectLatents:
+ """
+ Selects a range of frames from a video latent.
+
+ Features:
+ - Supports positive and negative indexing
+ - Preserves batch processing capabilities
+ - Handles noise masks if present
+ - Maintains 5D tensor format
+ """
+
+ @classmethod
+ def INPUT_TYPES(s):
+ return {
+ "required": {
+ "samples": ("LATENT",),
+ "start_index": (
+ "INT",
+ {"default": 0, "min": -9999, "max": 9999, "step": 1},
+ ),
+ "end_index": (
+ "INT",
+ {"default": -1, "min": -9999, "max": 9999, "step": 1},
+ ),
+ }
+ }
+
+ RETURN_TYPES = ("LATENT",)
+ FUNCTION = "select_latents"
+ CATEGORY = "latent/video"
+
+ def select_latents(self, samples: dict, start_index: int, end_index: int) -> tuple:
+ """
+ Selects a range of frames from the video latent.
+
+ Args:
+ samples (dict): Video latent dictionary
+ start_index (int): Starting frame index (supports negative indexing)
+ end_index (int): Ending frame index (supports negative indexing)
+
+ Returns:
+ tuple: Contains modified latent dictionary with selected frames
+
+ Raises:
+ ValueError: If indices are invalid
+ """
+ try:
+ s = samples.copy()
+ video_latent = s["samples"]
+ batch, channels, frames, height, width = video_latent.shape
+
+ # Handle negative indices
+ start_idx = frames + start_index if start_index < 0 else start_index
+ end_idx = frames + end_index if end_index < 0 else end_index
+
+ # Validate and clamp indices
+ start_idx = max(0, min(start_idx, frames - 1))
+ end_idx = max(0, min(end_idx, frames - 1))
+ if start_idx > end_idx:
+ start_idx = min(start_idx, end_idx)
+
+ # Select frames while maintaining 5D format
+ s["samples"] = video_latent[:, :, start_idx : end_idx + 1, :, :]
+
+ # Handle noise mask if present
+ if "noise_mask" in s:
+ s["noise_mask"] = s["noise_mask"][:, :, start_idx : end_idx + 1, :, :]
+
+ return (s,)
+
+ except Exception as e:
+ print(f"[LTXVSelectLatents] Error: {str(e)}")
+ raise
+
+
+@comfy_node(name="LTXVAddLatents")
+class LTXVAddLatents:
+ """
+ Concatenates two video latents along the frames dimension.
+
+ Features:
+ - Validates dimension compatibility
+ - Handles device placement
+ - Preserves noise masks with proper handling
+ - Supports batch processing
+ """
+
+ @classmethod
+ def INPUT_TYPES(s):
+ return {
+ "required": {
+ "latents1": ("LATENT",),
+ "latents2": ("LATENT",),
+ }
+ }
+
+ RETURN_TYPES = ("LATENT",)
+ FUNCTION = "add_latents"
+ CATEGORY = "latent/video"
+
+ def add_latents(
+ self, latents1: torch.Tensor, latents2: torch.Tensor
+ ) -> torch.Tensor:
+ """
+ Concatenates two video latents along the frames dimension.
+
+ Args:
+ latents1 (dict): First video latent dictionary
+ latents2 (dict): Second video latent dictionary
+
+ Returns:
+ tuple: Contains concatenated latent dictionary
+
+ Raises:
+ ValueError: If latent dimensions don't match
+ RuntimeError: If tensor operations fail
+ """
+ try:
+ s = latents1.copy()
+ video_latent1 = latents1["samples"]
+ video_latent2 = latents2["samples"]
+
+ # Ensure tensors are on the same device
+ target_device = video_latent1.device
+ video_latent2 = video_latent2.to(target_device)
+
+ # Validate dimensions
+ self._validate_dimensions(video_latent1, video_latent2)
+
+ # Concatenate along frames dimension
+ s["samples"] = torch.cat([video_latent1, video_latent2], dim=2)
+
+ # Handle noise masks
+ s["noise_mask"] = self._merge_noise_masks(
+ latents1, latents2, video_latent1.shape[2], video_latent2.shape[2]
+ )
+
+ return (s,)
+
+ except Exception as e:
+ print(f"[LTXVAddLatents] Error: {str(e)}")
+ raise
+
+ def _validate_dimensions(self, latent1: torch.Tensor, latent2: torch.Tensor):
+ """Validates that latent dimensions match except for frames."""
+ b1, c1, f1, h1, w1 = latent1.shape
+ b2, c2, f2, h2, w2 = latent2.shape
+
+ if not (b1 == b2 and c1 == c2 and h1 == h2 and w1 == w2):
+ raise ValueError(
+ f"Latent dimensions must match (except frames dimension).\n"
+ f"Got shapes {latent1.shape} and {latent2.shape}"
+ )
+
+ def _merge_noise_masks(
+ self, latents1: torch.Tensor, latents2: torch.Tensor, frames1: int, frames2: int
+ ) -> Optional[torch.Tensor]:
+ """Merges noise masks from both latents with proper handling."""
+ if "noise_mask" in latents1 and "noise_mask" in latents2:
+ return torch.cat([latents1["noise_mask"], latents2["noise_mask"]], dim=2)
+ elif "noise_mask" in latents1:
+ zeros = torch.zeros_like(latents1["noise_mask"][:, :, :frames2, :, :])
+ return torch.cat([latents1["noise_mask"], zeros], dim=2)
+ elif "noise_mask" in latents2:
+ zeros = torch.zeros_like(latents2["noise_mask"][:, :, :frames1, :, :])
+ return torch.cat([zeros, latents2["noise_mask"]], dim=2)
+ return None
+
+
+@comfy_node(name="LTXVSetVideoLatentNoiseMasks")
+class LTXVSetVideoLatentNoiseMasks:
+ """
+ Applies multiple masks to a video latent.
+
+ Features:
+ - Supports multiple input mask formats (2D, 3D, 4D)
+ - Automatically handles fewer masks than frames by reusing the last mask
+ - Resizes masks to match latent dimensions
+ - Preserves batch processing capabilities
+
+ Input Formats:
+ - 2D mask: Single mask [H, W]
+ - 3D mask: Multiple masks [M, H, W]
+ - 4D mask: Multiple masks with channels [M, C, H, W]
+ """
+
+ @classmethod
+ def INPUT_TYPES(s):
+ return {
+ "required": {
+ "samples": ("LATENT",),
+ "masks": ("MASK",),
+ }
+ }
+
+ RETURN_TYPES = ("LATENT",)
+ FUNCTION = "set_mask"
+ CATEGORY = "latent/video"
+
+ def set_mask(self, samples: dict, masks: torch.Tensor) -> tuple:
+ """
+ Applies masks to video latent frames.
+
+ Args:
+ samples (dict): Video latent dictionary containing 'samples' tensor
+ masks (torch.Tensor): Mask tensor in various possible formats
+ - 2D: [H, W] single mask
+ - 3D: [M, H, W] multiple masks
+ - 4D: [M, C, H, W] multiple masks with channels
+
+ Returns:
+ tuple: Contains modified latent dictionary with applied masks
+
+ Raises:
+ ValueError: If mask dimensions are unsupported
+ RuntimeError: If tensor operations fail
+ """
+ try:
+ s = samples.copy()
+ video_latent = s["samples"]
+ batch_size, channels, num_frames, height, width = video_latent.shape
+
+ # Initialize noise_mask if not present
+ if "noise_mask" not in s:
+ s["noise_mask"] = torch.zeros(
+ (batch_size, 1, num_frames, height, width),
+ dtype=video_latent.dtype,
+ device=video_latent.device,
+ )
+
+ # Process masks
+ masks_reshaped = self._reshape_masks(masks)
+ M = masks_reshaped.shape[0]
+ resized_masks = self._resize_masks(masks_reshaped, height, width)
+
+ # Apply masks efficiently
+ self._apply_masks(s["noise_mask"], resized_masks, num_frames, M)
+ return (s,)
+
+ except Exception as e:
+ print(f"[LTXVSetVideoLatentNoiseMasks] Error: {str(e)}")
+ raise
+
+ def _reshape_masks(self, masks: torch.Tensor) -> torch.Tensor:
+ """Reshapes input masks to consistent 4D format."""
+ original_shape = tuple(masks.shape)
+ ndims = masks.ndim
+
+ if ndims == 2:
+ return masks.unsqueeze(0).unsqueeze(0)
+ elif ndims == 3:
+ return masks.reshape(masks.shape[0], 1, masks.shape[1], masks.shape[2])
+ elif ndims == 4:
+ return masks.reshape(masks.shape[0], 1, masks.shape[2], masks.shape[3])
+ else:
+ raise ValueError(
+ f"Unsupported 'masks' dimension: {original_shape}. "
+ "Must be 2D (H,W), 3D (M,H,W), or 4D (M,C,H,W)."
+ )
+
+ def _resize_masks(
+ self, masks: torch.Tensor, height: int, width: int
+ ) -> torch.Tensor:
+ """Resizes all masks to match latent dimensions."""
+ return torch.nn.functional.interpolate(
+ masks, size=(height, width), mode="bilinear", align_corners=False
+ )
+
+ def _apply_masks(
+ self,
+ noise_mask: torch.Tensor,
+ resized_masks: torch.Tensor,
+ num_frames: int,
+ M: int,
+ ) -> None:
+ """Applies resized masks to all frames."""
+ for f in range(num_frames):
+ mask_idx = min(f, M - 1) # Reuse last mask if we run out
+ noise_mask[:, :, f] = resized_masks[mask_idx]
+
+
+@comfy_node(name="LTXVAddLatentGuide")
+class LTXVAddLatentGuide(nodes_lt.LTXVAddGuide):
+ @classmethod
+ def INPUT_TYPES(s):
+ return {
+ "required": {
+ "vae": ("VAE",),
+ "positive": ("CONDITIONING",),
+ "negative": ("CONDITIONING",),
+ "latent": ("LATENT",),
+ "guiding_latent": ("LATENT",),
+ "latent_idx": (
+ "INT",
+ {
+ "default": 0,
+ "min": -9999,
+ "max": 9999,
+ "step": 1,
+ "tooltip": "Latent index to start the conditioning at. Can be negative to"
+ "indicate that the conditioning is on the frames before the latent.",
+ },
+ ),
+ "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0}),
+ }
+ }
+
+ RETURN_TYPES = ("CONDITIONING", "CONDITIONING", "LATENT")
+ RETURN_NAMES = ("positive", "negative", "latent")
+
+ CATEGORY = "ltxtricks"
+ FUNCTION = "generate"
+
+ def generate(
+ self, vae, positive, negative, latent, guiding_latent, latent_idx, strength
+ ):
+ noise_mask = nodes_lt.get_noise_mask(latent)
+ latent = latent["samples"]
+ guiding_latent = guiding_latent["samples"]
+ scale_factors = vae.downscale_index_formula
+
+ if latent_idx <= 0:
+ frame_idx = latent_idx * scale_factors[0]
+ else:
+ frame_idx = 1 + (latent_idx - 1) * scale_factors[0]
+
+ positive, negative, latent, noise_mask = self.append_keyframe(
+ positive=positive,
+ negative=negative,
+ frame_idx=frame_idx,
+ latent_image=latent,
+ noise_mask=noise_mask,
+ guiding_latent=guiding_latent,
+ strength=strength,
+ scale_factors=scale_factors,
+ )
+
+ return (
+ positive,
+ negative,
+ {"samples": latent, "noise_mask": noise_mask},
+ )
diff --git a/pyproject.toml b/pyproject.toml
index ba6d54c..b6a70c6 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -16,7 +16,8 @@ dependencies = [
[project.optional-dependencies]
internal = [
"ltx-video@git+https://github.com/Lightricks/[email protected]",
- "av>=10.0.0"
+ "av>=10.0.0",
+ "q8-kernels>=0.0.2"
]
dev = [
"pre-commit>=4.0.1",
diff --git a/recurrent_sampler.py b/recurrent_sampler.py
new file mode 100644
index 0000000..450301d
--- /dev/null
+++ b/recurrent_sampler.py
@@ -0,0 +1,254 @@
+import torch
+from comfy_extras.nodes_custom_sampler import CFGGuider, SamplerCustomAdvanced
+from comfy_extras.nodes_lt import LTXVAddGuide, LTXVCropGuides
+
+from .latents import LTXVAddLatents, LTXVSelectLatents
+from .nodes_registry import comfy_node
+from .tricks import AddLatentGuideNode
+
+
+@comfy_node(description="Linear transition with overlap")
+class LinearOverlapLatentTransition:
+ @classmethod
+ def INPUT_TYPES(s):
+ return {
+ "required": {
+ "samples1": ("LATENT",),
+ "samples2": ("LATENT",),
+ "overlap": ("INT", {"default": 1, "min": 1, "max": 256}),
+ },
+ "optional": {
+ "axis": ("INT", {"default": 0}),
+ },
+ }
+
+ RETURN_TYPES = ("LATENT",)
+ FUNCTION = "process"
+
+ CATEGORY = "Lightricks/latent"
+
+ def get_subbatch(self, samples):
+ s = samples.copy()
+ samples = s["samples"]
+ return samples
+
+ def process(self, samples1, samples2, overlap, axis=0):
+ samples1 = self.get_subbatch(samples1)
+ samples2 = self.get_subbatch(samples2)
+
+ # Create transition coefficients
+ alpha = torch.linspace(1, 0, overlap + 2)[1:-1].to(samples1.device)
+
+ # Create shape for broadcasting based on the axis
+ shape = [1] * samples1.dim()
+ shape[axis] = alpha.size(0)
+ alpha = alpha.reshape(shape)
+
+ # Create slices for the overlap regions
+ slice_all = [slice(None)] * samples1.dim()
+ slice_overlap1 = slice_all.copy()
+ slice_overlap1[axis] = slice(-overlap, None)
+ slice_overlap2 = slice_all.copy()
+ slice_overlap2[axis] = slice(0, overlap)
+ slice_rest1 = slice_all.copy()
+ slice_rest1[axis] = slice(None, -overlap)
+ slice_rest2 = slice_all.copy()
+ slice_rest2[axis] = slice(overlap, None)
+
+ # Combine samples
+ parts = [
+ samples1[tuple(slice_rest1)],
+ alpha * samples1[tuple(slice_overlap1)]
+ + (1 - alpha) * samples2[tuple(slice_overlap2)],
+ samples2[tuple(slice_rest2)],
+ ]
+
+ combined_samples = torch.cat(parts, dim=axis)
+ combined_batch_index = torch.arange(0, combined_samples.shape[0])
+
+ return (
+ {
+ "samples": combined_samples,
+ "batch_index": combined_batch_index,
+ },
+ )
+
+
+@comfy_node(
+ name="LTXVRecurrentKSampler",
+)
+class LTXVRecurrentKSampler:
+
+ @classmethod
+ def INPUT_TYPES(s):
+ return {
+ "required": {
+ "model": ("MODEL",),
+ "vae": ("VAE",),
+ "noise": ("NOISE",),
+ "sampler": ("SAMPLER",),
+ "sigmas": ("SIGMAS",),
+ "latents": ("LATENT",),
+ "chunk_sizes": ("STRING", {"default": "3", "multiline": False}),
+ "overlaps": ("STRING", {"default": "1", "multiline": False}),
+ "positive": ("CONDITIONING",),
+ "negative": ("CONDITIONING",),
+ "input_image": ("IMAGE",),
+ "linear_blend_latents": ("BOOLEAN", {"default": True}),
+ "conditioning_strength": (
+ "FLOAT",
+ {"default": 1.0, "min": 0.0, "max": 2.0, "step": 0.01},
+ ),
+ },
+ "optional": {
+ "guider": ("GUIDER",),
+ },
+ }
+
+ RETURN_TYPES = (
+ "LATENT",
+ "LATENT",
+ )
+ RETURN_NAMES = (
+ "output",
+ "denoised_output",
+ )
+
+ FUNCTION = "sample"
+
+ CATEGORY = "sampling"
+
+ def sample(
+ self,
+ model,
+ vae,
+ noise,
+ sampler,
+ sigmas,
+ latents,
+ chunk_sizes,
+ overlaps,
+ positive,
+ negative,
+ input_image,
+ linear_blend_latents,
+ conditioning_strength,
+ guider=None,
+ ):
+ select_latents = LTXVSelectLatents().select_latents
+ add_latent_guide = AddLatentGuideNode().generate
+ add_latents = LTXVAddLatents().add_latents
+ positive_orig = positive.copy()
+ negative_orig = negative.copy()
+
+ # Parse chunk sizes and overlaps from strings
+ chunk_sizes = [int(x) for x in chunk_sizes.split(",")]
+ overlaps = [int(x) for x in overlaps.split(",")]
+
+ # Extend lists if shorter than number of sigma steps
+ n_steps = len(sigmas) - 1
+ if len(chunk_sizes) < n_steps:
+ chunk_sizes.extend([chunk_sizes[-1]] * (n_steps - len(chunk_sizes)))
+ if len(overlaps) < n_steps:
+ overlaps.extend([overlaps[-1]] * (n_steps - len(overlaps)))
+
+ # Initialize working latents
+ current_latents = latents.copy()
+ t_latents = None
+ # Loop through sigma pairs for progressive denoising
+ for i in range(n_steps):
+ current_sigmas = sigmas[i : i + 2]
+ current_chunk_size = chunk_sizes[i]
+ current_overlap = overlaps[i]
+
+ print(f"\nProcessing sigma step {i} with sigmas {current_sigmas}")
+ print(
+ f"Using chunk size {current_chunk_size} and overlap {current_overlap}"
+ )
+
+ # Calculate valid chunk starts to ensure the last chunk isn't shorter than the overlap
+ total_frames = current_latents["samples"].shape[2]
+ chunk_stride = current_chunk_size - current_overlap
+ valid_chunk_starts = list(
+ range(0, total_frames - current_overlap, chunk_stride)
+ )
+
+ # If the last chunk would be too short, remove the last start position
+ if (
+ total_frames > chunk_stride
+ and (total_frames - valid_chunk_starts[-1]) < current_chunk_size
+ ):
+ print(
+ "last chunk is too short, it will only be of size",
+ total_frames - valid_chunk_starts[-1],
+ "frames",
+ )
+
+ # Process each chunk for current sigma pair
+ for i_chunk, chunk_start in enumerate(valid_chunk_starts):
+ (latents_chunk,) = select_latents(
+ current_latents, chunk_start, chunk_start + current_chunk_size - 1
+ )
+ print(f"Processing chunk {i_chunk} starting at frame {chunk_start}")
+
+ if i_chunk == 0:
+ positive, negative, latents_chunk = LTXVAddGuide().generate(
+ positive_orig,
+ negative_orig,
+ vae,
+ latents_chunk,
+ input_image,
+ 0,
+ 0.75,
+ )
+ else:
+ (cond_latent,) = select_latents(t_latents, -current_overlap, -1)
+ model, latents_chunk = add_latent_guide(
+ model, latents_chunk, cond_latent, 0, conditioning_strength
+ )
+
+ if guider is None:
+ (guider_obj,) = CFGGuider().get_guider(
+ model, positive, negative, 1.0
+ )
+ else:
+ guider_obj = guider
+ (_, denoised_latents) = SamplerCustomAdvanced().sample(
+ noise, guider_obj, sampler, current_sigmas, latents_chunk
+ )
+ (positive, negative, denoised_latents) = LTXVCropGuides().crop(
+ positive, negative, denoised_latents
+ )
+
+ if i_chunk == 0:
+ t_latents = denoised_latents
+ else:
+ if linear_blend_latents and current_overlap > 1:
+ # the first output latent is the result of a 1:8 latent
+ # reinterpreted as a 1:1 latent, so we ignore it
+ (denoised_latents_drop_first,) = select_latents(
+ denoised_latents, 1, -1
+ )
+ (t_latents,) = LinearOverlapLatentTransition().process(
+ t_latents,
+ denoised_latents_drop_first,
+ current_overlap - 1,
+ axis=2,
+ )
+ else:
+ (truncated_denoised_latents,) = select_latents(
+ denoised_latents, current_overlap, -1
+ )
+ (t_latents,) = add_latents(
+ t_latents, truncated_denoised_latents
+ )
+
+ print(
+ f"Completed chunk {i_chunk}, current output shape: {t_latents['samples'].shape}"
+ )
+
+ # Update current_latents for next sigma step
+ current_latents = t_latents.copy()
+ print(f"Completed sigma step {i}")
+
+ return t_latents, t_latents
diff --git a/stg.py b/stg.py
index a7d18fe..b6f1250 100644
--- a/stg.py
+++ b/stg.py
@@ -1,10 +1,10 @@
import contextlib
+import math
from dataclasses import dataclass
from typing import List
import comfy.ldm.modules.attention
import comfy.samplers
-import comfy.utils
import torch
from comfy.model_patcher import ModelPatcher
@@ -20,8 +20,8 @@ def stg(
rescale_scale,
):
noise_pred = (
- noise_pred_neg
- + cfg_scale * (noise_pred_pos - noise_pred_neg)
+ noise_pred_pos
+ + (cfg_scale - 1) * (noise_pred_pos - noise_pred_neg)
+ stg_scale * (noise_pred_pos - noise_pred_pertubed)
)
if rescale_scale != 0:
@@ -183,6 +183,158 @@ class STGGuider(comfy.samplers.CFGGuider):
positive_cond = self.conds.get("positive", None)
negative_cond = self.conds.get("negative", None)
+ if model_options.get("sigma_to_params_mapping", None) is not None:
+ cfg_value, stg_scale, stg_layer_skip_layer_indices, stg_rescale = (
+ model_options["sigma_to_params_mapping"](timestep)
+ )
+ self.stg_flag.skip_layers = stg_layer_skip_layer_indices
+ self.patch_model(self.model_patcher, self.stg_flag)
+
+ else:
+ cfg_value = self.cfg
+ stg_scale = self.stg_scale
+ stg_rescale = self.rescale_scale
+
+ noise_pred_pos = comfy.samplers.calc_cond_batch(
+ self.inner_model,
+ [positive_cond],
+ x,
+ timestep,
+ model_options,
+ )[0]
+
+ noise_pred_neg = 0
+ noise_pred_perturbed = 0
+
+ if not math.isclose(cfg_value, 1.0):
+ noise_pred_neg = comfy.samplers.calc_cond_batch(
+ self.inner_model,
+ [negative_cond],
+ x,
+ timestep,
+ model_options,
+ )[0]
+
+ if not math.isclose(stg_scale, 0.0):
+ try:
+ model_options["transformer_options"]["ptb_index"] = 0
+ self.stg_flag.do_skip = True
+ noise_pred_perturbed = comfy.samplers.calc_cond_batch(
+ self.inner_model,
+ [positive_cond],
+ x,
+ timestep,
+ model_options,
+ )[0]
+ finally:
+ self.stg_flag.do_skip = False
+ del model_options["transformer_options"]["ptb_index"]
+
+ stg_result = stg(
+ noise_pred_pos,
+ noise_pred_neg,
+ noise_pred_perturbed,
+ cfg_value,
+ stg_scale,
+ stg_rescale,
+ )
+
+ # normally this would be done in cfg_function, but we skipped
+ # that for efficiency: we can compute the noise predictions in
+ # a single call to calc_cond_batch() (rather than two)
+ # so we replicate the hook here
+ for fn in model_options.get("sampler_post_cfg_function", []):
+ args = {
+ "denoised": stg_result,
+ "cond": positive_cond,
+ "uncond": negative_cond,
+ "model": self.inner_model,
+ "uncond_denoised": noise_pred_neg,
+ "cond_denoised": noise_pred_pos,
+ "sigma": timestep,
+ "model_options": model_options,
+ "input": x,
+ # not in the original call in samplers.py:cfg_function, but made available for future hooks
+ "perturbed_cond": positive_cond,
+ "perturbed_cond_denoised": noise_pred_perturbed,
+ }
+ stg_result = fn(args)
+
+ return stg_result
+
+
+class STGGuiderAdvanced(comfy.samplers.CFGGuider):
+ def __init__(
+ self,
+ model: ModelPatcher,
+ sigma_list,
+ cfg_list,
+ stg_scale_list,
+ stg_rescale_list,
+ stg_layers_indices_list,
+ skip_steps_sigma_threshold,
+ cfg_star_rescale,
+ ):
+ model = model.clone()
+ super().__init__(model)
+
+ self.stg_flag = STGFlag(
+ do_skip=False,
+ skip_layers=model.model_options["transformer_options"].get(
+ "skip_block_list"
+ ),
+ )
+
+ self.sigma_list = sigma_list
+ self.cfg_list = cfg_list
+ self.stg_scale_list = stg_scale_list
+ self.stg_rescale_list = stg_rescale_list
+ self.stg_layers_indices_list = stg_layers_indices_list
+ self.skip_steps_sigma_threshold = skip_steps_sigma_threshold
+ self.cfg_star_rescale = cfg_star_rescale
+ STGGuider.patch_model(model, self.stg_flag)
+
+ def sigma_to_params_mapping(self, sigma):
+ # Find the closest higher sigma value and return corresponding cfg
+ higher_sigmas = [s for s in self.sigma_list if s >= sigma]
+ if not higher_sigmas:
+ return self.cfg_list[-1] # Return last cfg if no higher sigma exists
+ closest_higher = min(higher_sigmas)
+ closest_idx = self.sigma_list.index(closest_higher)
+ return (
+ self.cfg_list[closest_idx],
+ self.stg_scale_list[closest_idx],
+ self.stg_rescale_list[closest_idx],
+ self.stg_layers_indices_list[closest_idx],
+ )
+
+ def set_conds(self, positive, negative):
+ self.inner_set_conds({"positive": positive, "negative": negative})
+
+ def predict_noise(
+ self,
+ x: torch.Tensor,
+ timestep: torch.Tensor,
+ model_options: dict = {},
+ seed=None,
+ ):
+ # CFG zero init - skipping steps with timestep bigger than given threshold.
+ if timestep > self.skip_steps_sigma_threshold:
+ return torch.zeros_like(x)
+
+ # in CFGGuider.predict_noise, we call sampling_function(), which uses cfg_function() to compute pos & neg
+ # but we'd rather do a single batch of sampling pos, neg, and perturbed, so we call calc_cond_batch([perturbed,pos,neg]) directly
+ positive_cond = self.conds.get("positive", None)
+ negative_cond = self.conds.get("negative", None)
+
+ cfg_value, stg_scale, stg_rescale, stg_layer_skip_layer_indices = (
+ self.sigma_to_params_mapping(timestep)
+ )
+
+ if stg_layer_skip_layer_indices is not None:
+ self.stg_flag.skip_layers = stg_layer_skip_layer_indices
+ STGGuider.patch_model(self.model_patcher, self.stg_flag)
+
noise_pred_pos = comfy.samplers.calc_cond_batch(
self.inner_model,
[positive_cond],
@@ -194,7 +346,7 @@ class STGGuider(comfy.samplers.CFGGuider):
noise_pred_neg = 0
noise_pred_perturbed = 0
- if self.cfg > 1:
+ if not math.isclose(cfg_value, 1.0):
noise_pred_neg = comfy.samplers.calc_cond_batch(
self.inner_model,
[negative_cond],
@@ -203,7 +355,19 @@ class STGGuider(comfy.samplers.CFGGuider):
model_options,
)[0]
- if self.stg_scale > 0:
+ if self.cfg_star_rescale:
+ batch_size = noise_pred_pos.shape[0]
+
+ positive_flat = noise_pred_pos.view(batch_size, -1)
+ negative_flat = noise_pred_neg.view(batch_size, -1)
+ dot_product = torch.sum(
+ positive_flat * negative_flat, dim=1, keepdim=True
+ )
+ squared_norm = torch.sum(negative_flat**2, dim=1, keepdim=True) + 1e-8
+ alpha = dot_product / squared_norm
+ noise_pred_neg = alpha * noise_pred_neg
+
+ if not math.isclose(stg_scale, 0.0):
try:
model_options["transformer_options"]["ptb_index"] = 0
self.stg_flag.do_skip = True
@@ -222,9 +386,9 @@ class STGGuider(comfy.samplers.CFGGuider):
noise_pred_pos,
noise_pred_neg,
noise_pred_perturbed,
- self.cfg,
- self.stg_scale,
- self.rescale_scale,
+ cfg_value,
+ stg_scale,
+ stg_rescale,
)
# normally this would be done in cfg_function, but we skipped
@@ -290,3 +454,130 @@ class STGGuiderNode:
guider = STGGuider(model, cfg, stg, rescale)
guider.set_conds(positive, negative)
return (guider,)
+
+
+@comfy_node(name="STGGuiderAdvanced")
+class STGGuiderAdvancedNode:
+ @classmethod
+ def INPUT_TYPES(s):
+ return {
+ "required": {
+ "model": ("MODEL",),
+ "positive": ("CONDITIONING",),
+ "negative": ("CONDITIONING",),
+ "skip_steps_sigma_threshold": (
+ "FLOAT",
+ {
+ "default": 0.996,
+ "min": 0.0,
+ "max": 100.0,
+ "step": 0.001,
+ "tooltip": "Steps with sigma greater than this values will be skipped.",
+ },
+ ),
+ "cfg_star_rescale": (
+ "BOOLEAN",
+ {
+ "default": True,
+ "tooltip": "If true, applies the cfg star rescale, scales the negative prediction according to dot product between positive and negative.",
+ },
+ ),
+ "sigmas": (
+ "STRING",
+ {
+ "default": "1.0, 0.9933, 0.9850, 0.9767, 0.9008, 0.6180",
+ "tooltip": "Comma-separated list sigmas, the actual parameters will be selected according to the closest sigma from this list to current timestep sigma.",
+ },
+ ),
+ "cfg_values": (
+ "STRING",
+ {
+ "default": "4, 4, 4, 4, 1, 1",
+ "tooltip": "Comma-separated list of cfg values. Should be same length as sigmas list.",
+ },
+ ),
+ "stg_scale_values": (
+ "STRING",
+ {
+ "default": "2, 2, 2, 2, 1, 0",
+ "tooltip": "Comma-separated list of stg scale values. Should be same length as sigmas list.",
+ },
+ ),
+ "stg_rescale_values": (
+ "STRING",
+ {
+ "default": "1, 1, 1, 1, 1, 1",
+ "tooltip": "Comma-separated list of stg rescale values. Should be same length as sigmas list.",
+ },
+ ),
+ "stg_layers_indices": (
+ "STRING",
+ {
+ "default": "[14], [14], [14], [14], [14], [14]",
+ "tooltip": "Comma-separated list of list of layer indices. Should be same length as sigmas list.",
+ },
+ ),
+ }
+ }
+
+ RETURN_TYPES = ("GUIDER",)
+
+ FUNCTION = "get_guider"
+ CATEGORY = "lightricks/LTXV"
+
+ @classmethod
+ def parse_stg_layers_indices(cls, stg_layers_indices: str) -> List[List[int]]:
+ # First split by "], " to separate the lists, but preserve the closing bracket
+ lists = [s + "]" for s in stg_layers_indices.split("],")[:-1]]
+ # Add the last list which already has its closing bracket
+ if stg_layers_indices.strip():
+ lists.append(stg_layers_indices.split("],")[-1])
+
+ result = []
+ for s in lists:
+ s = s.strip()
+ if s == "[]": # Empty list case
+ result.append([])
+ else:
+ # Remove brackets and whitespace
+ s = s.strip("[]").strip()
+ if not s: # Handle case like "[]" after split
+ result.append([])
+ else:
+ # Split by comma and convert to integers
+ numbers = [int(n.strip()) for n in s.split(",") if n.strip()]
+ result.append(numbers)
+
+ return result
+
+ def get_guider(
+ self,
+ model,
+ positive,
+ negative,
+ skip_steps_sigma_threshold,
+ cfg_star_rescale,
+ sigmas,
+ cfg_values,
+ stg_scale_values,
+ stg_rescale_values,
+ stg_layers_indices,
+ ):
+ sigma_list = [float(s.strip()) for s in sigmas.split(",")]
+ cfg_list = [float(c.strip()) for c in cfg_values.split(",")]
+ stg_scale_list = [float(s.strip()) for s in stg_scale_values.split(",")]
+ stg_rescale_list = [float(s.strip()) for s in stg_rescale_values.split(",")]
+ stg_layers_indices_list = self.parse_stg_layers_indices(stg_layers_indices)
+
+ guider = STGGuiderAdvanced(
+ model,
+ sigma_list,
+ cfg_list,
+ stg_scale_list,
+ stg_rescale_list,
+ stg_layers_indices_list,
+ skip_steps_sigma_threshold,
+ cfg_star_rescale,
+ )
+ guider.set_conds(positive, negative)
+ return (guider,)
diff --git a/tricks/assets/ltxvideo-flow-edit.json b/tricks/assets/ltxvideo-flow-edit.json
index c45b66b..c0af777 100644
--- a/tricks/assets/ltxvideo-flow-edit.json
+++ b/tricks/assets/ltxvideo-flow-edit.json
@@ -1302,7 +1302,7 @@
"Node name for S&R": "CheckpointLoaderSimple"
},
"widgets_values": [
- "ltx-video-2b-v0.9.5.safetensors"
+ "ltxv-2b-0.9.6-dev-04-25.safetensors"
]
},
{
@@ -2547,7 +2547,7 @@
},
"44": {
"inputs": {
- "ckpt_name": "ltx-video-2b-v0.9.5.safetensors"
+ "ckpt_name": "ltxv-2b-0.9.6-dev-04-25.safetensors"
},
"class_type": "CheckpointLoaderSimple",
"_meta": {
diff --git a/tricks/assets/ltxvideo-rf-edit.json b/tricks/assets/ltxvideo-rf-edit.json
index c0ceb5f..91eed8e 100644
--- a/tricks/assets/ltxvideo-rf-edit.json
+++ b/tricks/assets/ltxvideo-rf-edit.json
@@ -1574,7 +1574,7 @@
"Node name for S&R": "CheckpointLoaderSimple"
},
"widgets_values": [
- "ltx-video-2b-v0.9.5.safetensors"
+ "ltxv-2b-0.9.6-dev-04-25.safetensors"
]
},
{
@@ -3362,7 +3362,7 @@
},
"44": {
"inputs": {
- "ckpt_name": "ltx-video-2b-v0.9.5.safetensors"
+ "ckpt_name": "ltxv-2b-0.9.6-dev-04-25.safetensors"
},
"class_type": "CheckpointLoaderSimple",
"_meta": {
diff --git a/tricks/nodes/latent_guide_node.py b/tricks/nodes/latent_guide_node.py
index d3d1bc2..652a689 100644
--- a/tricks/nodes/latent_guide_node.py
+++ b/tricks/nodes/latent_guide_node.py
@@ -10,6 +10,7 @@ class AddLatentGuideNode(nodes_lt.LTXVAddGuide):
"latent": ("LATENT",),
"image_latent": ("LATENT",),
"index": ("INT", {"default": 0, "min": -1, "max": 9999, "step": 1}),
+ "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0}),
}
}
@@ -19,7 +20,7 @@ class AddLatentGuideNode(nodes_lt.LTXVAddGuide):
CATEGORY = "ltxtricks"
FUNCTION = "generate"
- def generate(self, model, latent, image_latent, index):
+ def generate(self, model, latent, image_latent, index, strength):
noise_mask = nodes_lt.get_noise_mask(latent)
latent = latent["samples"]
@@ -30,7 +31,7 @@ class AddLatentGuideNode(nodes_lt.LTXVAddGuide):
noise_mask,
image_latent,
index,
- 1.0,
+ strength,
)
return (