Add shoe first/last-frame example + workflow, workflows README and input frames; update model card, prompt guide and example workflow

README.md

@@ -41,31 +41,31 @@ widget:
     example_title: "Fighter jet · COMPLEX"
     output:
       url: examples/xpldv_jet.mp4
+  - text: "A premium running shoe separates into an ordered exploded view. The woven upper and laces lift together as one recognizable shoe-shaped shell. The insole rises beneath the upper, the pale cushioning midsole separates into its own curved horizontal layer, and the thin carbon-fiber plate lifts slightly above the dark rubber outsole. Each long shoe-shaped component stays aligned heel to toe, revealing the cushioning structure and support plate between the upper and the sole. The outsole remains on the display surface. The parts settle into a clean suspended stack with distinct air gaps and retain their original shapes. The camera slowly arcs to the right and pulls back enough to keep the full arrangement visible. Cool teal studio light and a restrained warm edge highlight remain consistent throughout the shot."
+    example_title: "Running shoe · First + last frame"
+    output:
+      url: examples/xpldv_shoe_first_last.mp4
 ---
 
 # XPLDV — Exploded-View LoRA for LTX-2.5
 
 Turn a source image into an animated exploded view. XPLDV separates an object's exterior parts to reveal the components underneath, for product showcases, mechanical reveals, architectural breakdowns, and stylized cutaways.
 
-[Download LoRA](ltx-2.5-xpldv_v5-lora.safetensors) · [ComfyUI workflow](workflows/ltx-2.5_distilled_xpldv_example_workflow.json) · [Prompt guide](prompt_guide.md)
+[Download LoRA](ltx-2.5-xpldv_v5-lora.safetensors) · [ComfyUI workflow](workflows/ltx-2.5_distilled_xpldv_example_workflow.json) · [Sample image](workflows/xpldv_computer.png) · [Workflow guide](workflows/README.md) · [Prompt guide](prompt_guide.md)
 
 ## How it works
 
-1. Load the included workflow and select your starting image and installed model files.
-2. Describe which parts separate, what they reveal, and how the camera moves.
+1. Load the included workflow and select your installed model files.
+2. Download the [sample PC image](workflows/xpldv_computer.png) and upload it in **Starting image**, or copy it to `ComfyUI/input/xpldv_computer.png`. The workflow is prefilled with its exact tested prompt.
 3. Generate the video with XPLDV at strength **0.5**.
 
-XPLDV works best with `ltx-2.5-22b-distilled-transformer-bf16.safetensors`. It can also be used with `ltx-2.5-22b-dev-transformer-bf16.safetensors`, which it was trained on. The included workflow uses the distilled model. Base models: [Lightricks/LTX-2.5](https://huggingface.co/Lightricks/LTX-2.5).
-
-## First and last frame guidance
-
-First/last-frame generation also works well with XPLDV. Supply the assembled subject as the first image and a matching exploded arrangement as the last image. **In this setup, XPLDV provides motion guidance for the separation between the two images.** The endpoint images establish the starting appearance and intended final arrangement; the prompt describes which parts detach, their paths, and the ongoing subject and camera movement.
+For your own subject, replace the image and fill the BASIC template in the note directly above the positive prompt, or use the COMPLEX template in the reference note. Describe which parts separate, what they reveal, and how the camera moves.
 
-This approach worked well for the burger and running-shoe reveals in the demo reel. Use a workflow that accepts both image guides. The included example workflow uses one starting image; see the [first/last-frame setup and prompting notes](prompt_guide.md#first-and-last-frame-guidance) for the tested two-guide setup.
+The recommended starting model, based on our tests, is `ltx-2.5-22b-distilled-transformer-bf16.safetensors`. It can also be used with `ltx-2.5-22b-dev-transformer-bf16.safetensors`, which it was trained on. The included workflow uses the distilled model. Base models: [Lightricks/LTX-2.5](https://huggingface.co/Lightricks/LTX-2.5).
 
 ## Featured videos
 
-Six 10-second examples using XPLDV at strength **0.5**. Each video includes its exact prompt.
+Six 10-second single-image examples at strength **0.5**, plus a six-second shoe example at **0.7** using first and last frame guidance. Each video includes its exact recorded prompt.
 
 <Gallery />
 
@@ -74,7 +74,9 @@ Six 10-second examples using XPLDV at strength **0.5**. Each video includes its
 
 Matched comparisons using the same image, prompt, seed, and settings. Comparison videos are silent.
 
-[Laptop](examples/comparisons/comparison_laptop.mp4) · [Computer](examples/comparisons/comparison_computer.mp4) · [SUV](examples/comparisons/comparison_suv.mp4) · [Watch](examples/comparisons/comparison_watch.mp4) · [House](examples/comparisons/comparison_house.mp4) · [Fighter jet](examples/comparisons/comparison_jet.mp4)
+[Laptop](examples/comparisons/comparison_laptop.mp4) · [Computer](examples/comparisons/comparison_computer.mp4) · [SUV](examples/comparisons/comparison_suv.mp4) · [Watch](examples/comparisons/comparison_watch.mp4) · [House](examples/comparisons/comparison_house.mp4) · [Fighter jet](examples/comparisons/comparison_jet.mp4) · [Running shoe](examples/comparisons/comparison_shoe.mp4)
+
+The shoe comparison keeps both image guides fixed and uses the recorded strength of **0.7** on the XPLDV side. Compare the movement and component behavior between the guided endpoints. [Shoe workflow and guidance](workflows/README.md#first-and-last-frame-guidance).
 
 </details>
 
@@ -90,12 +92,14 @@ Matched comparisons using the same image, prompt, seed, and settings. Comparison
 | Second stage | 1920 × 1088; Euler; 3 steps |
 | Second-stage manual sigmas | `0.85, 0.7250, 0.4219, 0.0` |
 | Image conditioning | 0.7 in the first stage; 1.0 in the second |
-| Prompt enhancement | Off |
+| Prompt enhancement | Can be enabled; off in the recorded examples |
 | Negative prompt | Empty |
 | Seed for comparisons | 42 in both stages |
 
 The workflow includes these settings and applies the LoRA to both stages. The manual sigmas above belong in **Upscale and re-sampler (3 steps)**; keep the first-stage schedule unchanged.
 
+**Prompt enhancement:** Enabling **Enhance prompt** works with XPLDV, but results may vary. It has not been extensively tested. The recorded examples used enhancement off to preserve the supplied prompts.
+
 <details>
 <summary>LoRA strength</summary>
 
@@ -129,7 +133,7 @@ Use the [slot reference](prompt_guide.md#filling-the-slots) to fill the placehol
 The desktop computer tower separates into an ordered exploded view. The glass side panel, front panel, top cover and lower power-supply cover move outward along controlled paths, revealing the motherboard, graphics card, liquid-cooling assembly and power supply. The components separate into an organized layered arrangement around the case while retaining their original shapes. The case base stays on the desk. The camera slowly orbits to the right around the computer from the starting three-quarter view. The blue-white computer lighting and surrounding room remain consistent throughout the continuous shot.
 ```
 
-[Watch the example](examples/xpldv_computer.mp4)
+[Watch the example](examples/xpldv_computer.mp4) · [Download its starting image](workflows/xpldv_computer.png)
 
 ### COMPLEX template — a detailed, cinematic reveal
 

examples/comparisons/comparison_shoe.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fe69b37725e54f47a658747f18b2f96d618ed2fa75fb6e50ae46c1bdcf8351c
+size 2736029

examples/xpldv_shoe_first_last.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6964930b4d161af74674234ab4336d31d309015318c77088d8084611ef3f86ec
+size 2177838

prompt_guide.md

@@ -1,23 +1,35 @@
 # Prompting XPLDV
 
-Start with **“separates into an ordered exploded view.”** Choose a [BASIC or COMPLEX template](README.md#prompting), then use the slot reference below to fill it.
+Start with **“separates into an ordered exploded view.”** Choose a [BASIC or COMPLEX template](#build-the-reveal), then use the slot reference below to fill it.
+
+**Prompt enhancement:** Enabling **Enhance prompt** works with XPLDV, but results may vary. It has not been extensively tested.
 
 ## Build the reveal
 
+### BASIC template — a short, readable reveal
+
+```text
+[SUBJECT] separates into an ordered exploded view. [NAMED OUTER PARTS] move outward along controlled paths, revealing [NAMED INTERNAL COMPONENTS] underneath. The parts settle into an organized [LAYERED / RADIAL] arrangement around [EXPOSED CORE] while retaining their original shapes. [CAMERA BEHAVIOR]. [LIGHTING CONSISTENT WITH THE IMAGE] remains consistent throughout the continuous shot.
+```
+
+### COMPLEX template — a detailed, cinematic reveal
+
+```text
+[SUBJECT] separates into an ordered exploded view while [CONTINUING ACTION OR ANCHORED SUPPORT]. [OUTER PART A] detaches and translates [DIRECTION], keeping [ORIENTATION]. [OUTER PART B] moves [DISTINCT PATH], and [OTHER OUTER PARTS] move [PATHS]. These exterior parts move first and travel furthest, leaving clear gaps. They reveal [SPECIFIC COMPONENT AND RECOGNIZABLE FEATURES] at [POSITION RELATIVE TO THE OBJECT], with [OTHER COMPONENTS AND THEIR LOCATIONS]. [CORE COMPONENTS] remain mounted together. Each detached section retains its shape as the arrangement settles and holds steady. [CAMERA BEHAVIOR]. [CONSISTENT LIGHTING] continues through one continuous shot.
+```
+
 ### Filling the slots
 
 Grouped names share an explanation. Fill only the slots in your chosen template, giving each part its own details, then write the finished prompt as one paragraph.
 
+#### XPLDV-specific separation details
+
 | Slot | What goes in it |
 | --- | --- |
-| `[SUBJECT]` | The thing being taken apart: “the black SUV.” |
 | `[NAMED OUTER PARTS]`<br>`[OUTER PART A]`<br>`[OUTER PART B]`<br>`[OTHER OUTER PARTS]` | Outside pieces to detach: “hood, front bumper, fenders.” Use the individual part slots when filling COMPLEX. |
 | `[NAMED INTERNAL COMPONENTS]` | Inside parts to reveal: “engine, radiator, and front seats.” |
 | `[LAYERED / RADIAL]` | Stacked layers or parts around a center. Example: “radial.” |
 | `[EXPOSED CORE]` | The center the parts surround: “the moving chassis.” |
-| `[CAMERA BEHAVIOR]` | One camera action with room for all the parts: “The camera tracks alongside at matching speed.” |
-| `[LIGHTING CONSISTENT WITH THE IMAGE]`<br>`[CONSISTENT LIGHTING]` | The light already in the image: “the warm street lighting.” |
-| `[CONTINUING ACTION OR ANCHORED SUPPORT]` | What the subject keeps doing or rests on: “driving across the bridge.” |
 | `[DIRECTION]`<br>`[DISTINCT PATH]`<br>`[PATHS]` | Each part's movement path: “hood straight upward; bumper forward.” Give each part its own path. |
 | `[ORIENTATION]` | How it stays angled: “flat and horizontal.” |
 | `[SPECIFIC COMPONENT AND RECOGNIZABLE FEATURES]` | An internal part and its appearance: “a V8 engine with exposed intake pipes.” |
@@ -25,13 +37,24 @@ Grouped names share an explanation. Fill only the slots in your chosen template,
 | `[OTHER COMPONENTS AND THEIR LOCATIONS]` | Other revealed assemblies and their positions: “brake discs inside the wheels.” |
 | `[CORE COMPONENTS]` | Parts to keep assembled: “the engine and transmission.” Omit this clause if those parts should separate too. |
 
+Match the paths to the final arrangement: vertically apart for a stack, sideways for a row, or outward around the core for a radial spread. Describe one consistent order, whether parts separate together or in stages. Detached parts should keep their shapes and remain suspended with visible gaps.
+
+#### General LTX prompting
+
+These describe the shot whether or not XPLDV is loaded. The image supplies the scene appearance; the prompt directs the action, camera movement, and any requested audio. See [LTX’s image-to-video guide](https://docs.ltx.io/open-source-model/usage-guides/image-to-video).
+
+| Slot | What goes in it |
+| --- | --- |
+| `[SUBJECT]` | The thing being taken apart: “the black SUV.” |
+| `[CAMERA BEHAVIOR]` | One camera action with room for all the parts: “The camera tracks alongside at matching speed.” |
+| `[LIGHTING CONSISTENT WITH THE IMAGE]`<br>`[CONSISTENT LIGHTING]` | The light already in the image: “the warm street lighting.” |
+| `[CONTINUING ACTION OR ANCHORED SUPPORT]` | What the subject keeps doing or rests on: “driving across the bridge.” |
+
 ## Keep the shot consistent
 
 State ongoing motion separately from separation: the vehicle keeps driving, its wheels keep rotating, and the detached parts keep traveling with it after the gaps stop increasing. Keep stationary subjects on their support.
 
-Match the paths to the final arrangement: vertically apart for a stack, sideways for a row, or outward around the core for a radial spread. Describe one consistent order, whether parts separate together or in stages. Detached parts should keep their shapes and remain suspended with visible gaps.
-
-**Optional additions**
+**Scene preservation and audio**
 
 - **Preservation:** Name existing people, objects, or surroundings that should stay consistent. For example: “The driver remains seated and the bridge stays intact.” These are preservation requests, not guaranteed locks.
 - **Audio:** With audio generation enabled, add one final sentence for sound effects, ambience, music, or dialogue. Put spoken dialogue in quotation marks. Audio wording can influence the visible action, so review picture and sound together. See [LTX's prompting guide](https://ltx.io/blog/ltx-2-5-prompt-guide) for examples.
@@ -41,11 +64,11 @@ Match the paths to the final arrangement: vertically apart for a stack, sideways
 XPLDV also works well with first/last-frame generation. **For this setup, use XPLDV as motion guidance between the two endpoint images.** The first image establishes the assembled subject; the last image defines the intended exploded arrangement. XPLDV encourages ordered component separation during the movement between them.
 
 1. **Prepare a compatible image pair.** Keep the same subject, identifiable parts, materials, and lighting. Give the last image clear gaps between components and enough room for the full arrangement. Any change in viewpoint should fit the camera movement you request.
-2. **Use a workflow that accepts both guides.** The tested ComfyUI setup uses `LTXVAddGuide` at frame `0` and frame `-1`, with both guides applied in each sampling pass. The included release workflow is a single-image example; it needs a first/last-frame variant for this setup.
+2. **Use a workflow that accepts both guides.** The tested ComfyUI setup uses `LTXVAddGuide` at frame `0` and frame `-1`, with both guides applied in each sampling pass. Use the included [shoe first/last-frame workflow](workflows/ltx-2.5_distilled_xpldv_first_last_shoe.json), prefilled with its tested prompt, [first image](workflows/xpldv_shoe_first.png), and [last image](workflows/xpldv_shoe_last.png). See the [loading steps and recorded result](workflows/README.md#first-and-last-frame-guidance).
 3. **Describe the movement between the images.** Use the existing BASIC or COMPLEX structure. Name the parts, their detachment paths, what becomes visible, and what continues moving. Write a consistent route to the supplied final arrangement.
 4. **Review the whole transition.** Check component identity, detachment, and continuous movement between the endpoints. A matching last frame alone does not establish a clean exploded-view animation.
 
-An end image can strongly determine the final geometry. To judge XPLDV's contribution, compare the movement between endpoints with the LoRA enabled and disabled while keeping both images, prompt, seed, and sampling settings fixed.
+An end image can strongly determine the final geometry. To judge XPLDV's contribution, compare the movement between endpoints with the LoRA enabled and disabled while keeping both images, prompt, seed, and sampling settings fixed. In the guided burger and running-shoe controls, recognizable separated arrangements also formed with XPLDV disabled. Judge the LoRA by the motion and component behavior between the images, not by reaching the end image alone.
 
 <details>
 <summary>Tested first/last-frame settings</summary>
@@ -123,4 +146,3 @@ Before returning, check that the completed prompt keeps the chosen subject and s
 
 Return only the completed positive prompt as one continuous paragraph, without headings, alternatives, placeholders or explanations.
 ```
-

workflows/README.md

@@ -0,0 +1,23 @@
+# ComfyUI workflows
+
+[Single-image PC workflow](ltx-2.5_distilled_xpldv_example_workflow.json) · [PC starting image](xpldv_computer.png) · [Recommended settings](../README.md#recommended-settings) · [Prompt guide](../prompt_guide.md)
+
+## First and last frame guidance
+
+First/last-frame generation also works well with XPLDV. Supply the assembled subject as the first image and a matching exploded arrangement as the last image. **In this setup, XPLDV provides motion guidance for the separation between the two images.** The endpoint images establish the starting appearance and intended final arrangement; the prompt describes which parts detach, their paths, and the ongoing subject and camera movement.
+
+Try the running-shoe example from the demo reel. The workflow includes its exact tested prompt and applies both image guides in both sampling passes.
+
+| First image: assembled | Last image: exploded |
+| --- | --- |
+| ![Assembled running shoe](xpldv_shoe_first.png) | ![Exploded running shoe](xpldv_shoe_last.png) |
+
+1. Download the [shoe first/last-frame workflow](ltx-2.5_distilled_xpldv_first_last_shoe.json) and open it in ComfyUI.
+2. Download the [first image](xpldv_shoe_first.png) and [last image](xpldv_shoe_last.png). Upload them in the matching image nodes, or copy both files into `ComfyUI/input`.
+3. Select your installed model files and the released XPLDV LoRA. Keep the prefilled prompt and settings, then run the workflow.
+
+This example uses XPLDV at **0.7**, seed **42**, and **145 frames at 24 FPS** (about six seconds), with a two-pass 960 × 544 → 1920 × 1088 render. Each endpoint guide is set to **1.0**. The exact prompt is filled in; the reusable BASIC template sits in a note above it.
+
+[Watch the recorded shoe result](../examples/xpldv_shoe_first_last.mp4) · [First/last-frame prompting notes](../prompt_guide.md#first-and-last-frame-guidance)
+
+The recorded result includes generated audio. Its end image guides the final stack, with some foam reshaping during separation. The single-image PC workflow above remains the simpler starting point at strength **0.5**.

workflows/ltx-2.5_distilled_xpldv_example_workflow.json

@@ -1,7 +1,7 @@
 {
   "id": "394ed254-7306-42a2-9ae6-aa880ce4456d",
-  "revision": 0,
-  "last_node_id": 5579,
+  "revision": 1,
+  "last_node_id": 5580,
   "last_link_id": 13841,
   "nodes": [
     {
@@ -111,7 +111,16 @@
         544,
         18,
         0.7
-      ]
+      ],
+      "widgets_values_named": {
+        "length": 121,
+        "a": 0,
+        "bypass": false,
+        "width": 960,
+        "height": 544,
+        "img_compression": 18,
+        "strength": 0.7
+      }
     },
     {
       "id": 5532,
@@ -148,7 +157,7 @@
       ],
       "size": [
         570,
-        190
+        420
       ],
       "flags": {
         "collapsed": false
@@ -160,7 +169,7 @@
       "title": "XPLDV - Settings",
       "properties": {},
       "widgets_values": [
-        "## Inputs\n\nChoose a source image and keep **use image input** enabled for the tested recipe. Replace the bracketed fields in the positive prompt, or use the complete COMPLEX template in the XPLDV note.\n\nKeep **enhance positive prompt OFF** to use your exact wording. Leave the API key empty for local encoding.\n\nThe duration is 10 seconds at 24 FPS; the workflow computes 241 frames. Optional audio direction can be appended at the end of either prompt. Keep the audio VAE connected.\n\n\nModel URL: https://huggingface.co/DigitalByte/LTX-2.5-Exploded-View-XPLDV/blob/main/ltx-2.5-xpldv_v5-lora.safetensors\n"
+        "## Start with the included PC example\n\n1. Select your installed model files and XPLDV weights.\n2. Download `workflows/xpldv_computer.png` from the release and upload it in **Starting image**. You can also copy it into `ComfyUI/input/` with that filename. A workflow JSON does not contain the image pixels.\n3. The positive prompt already contains the exact tested PC prompt. Keep **use image input** enabled, **enhance positive prompt** off, and XPLDV strength at **0.5**.\n4. Generate the 10-second example. For another subject, replace the image and use the template note directly above the prompt.\n\nThe workflow computes 241 frames at 24 FPS. Leave the API key empty for local encoding. Keep the audio VAE connected; audio direction can be added at the end of the prompt.\n\n[Download sample image](https://huggingface.co/DigitalByte/LTX-2.5-Exploded-View-XPLDV/resolve/main/workflows/xpldv_computer.png)"
       ],
       "color": "#322",
       "bgcolor": "#533"
@@ -296,7 +305,12 @@
         42,
         1,
         "euler_ancestral"
-      ]
+      ],
+      "widgets_values_named": {
+        "noise_seed": 42,
+        "cfg": 1,
+        "sampler_name": "euler_ancestral"
+      }
     },
     {
       "id": 5531,
@@ -391,14 +405,22 @@
         64,
         128,
         32
-      ]
+      ],
+      "widgets_values_named": {
+        "fps": 30,
+        "bit_depth": 8,
+        "tile_size": 512,
+        "overlap": 64,
+        "temporal_size": 128,
+        "temporal_overlap": 32
+      }
     },
     {
       "id": 5509,
       "type": "PrimitiveStringMultiline",
       "pos": [
         -3570,
-        3450
+        3550
       ],
       "size": [
         350,
@@ -434,7 +456,7 @@
       "type": "PrimitiveFloat",
       "pos": [
         -3570,
-        3960
+        4070
       ],
       "size": [
         350,
@@ -508,7 +530,7 @@
       "type": "LoadImage",
       "pos": [
         -3570,
-        3600
+        3700
       ],
       "size": [
         350,
@@ -538,11 +560,16 @@
         "Node name for S&R": "LoadImage"
       },
       "widgets_values": [
-        "",
+        "xpldv_computer.png",
         "image"
       ],
       "color": "#232",
-      "bgcolor": "#353"
+      "bgcolor": "#353",
+      "title": "Starting image - xpldv_computer.png",
+      "widgets_values_named": {
+        "image": "xpldv_computer.png",
+        "upload": "image"
+      }
     },
     {
       "id": 5014,
@@ -737,7 +764,18 @@
         457289196,
         "",
         "ltx-2.5-22b-distilled-transformer-bf16.safetensors"
-      ]
+      ],
+      "widgets_values_named": {
+        "value_3": true,
+        "value_2": "",
+        "cond": false,
+        "text": "",
+        "value": 24,
+        "value_5": 10,
+        "seed": 457289196,
+        "value_4": "",
+        "ckpt_name": "ltx-2.5-22b-distilled-transformer-bf16.safetensors"
+      }
     },
     {
       "id": 5530,
@@ -774,7 +812,7 @@
       ],
       "size": [
         350,
-        200
+        300
       ],
       "flags": {},
       "order": 9,
@@ -789,14 +827,14 @@
           ]
         }
       ],
-      "title": "Prompt (positive)",
+      "title": "Positive prompt - Desktop PC example",
       "properties": {
         "cnr_id": "comfy-core",
         "ver": "0.25.0",
         "Node name for S&R": "PrimitiveStringMultiline"
       },
       "widgets_values": [
-        "[SUBJECT] separates into an ordered exploded view. [NAMED OUTER PARTS] move outward along controlled paths, revealing [NAMED INTERNAL COMPONENTS] underneath. The parts settle into an organized [LAYERED / RADIAL] arrangement around [EXPOSED CORE] while retaining their original shapes. [CAMERA BEHAVIOR]. [LIGHTING CONSISTENT WITH THE IMAGE] remains consistent throughout the continuous shot."
+        "The desktop computer tower separates into an ordered exploded view. The glass side panel, front panel, top cover and lower power-supply cover move outward along controlled paths, revealing the motherboard, graphics card, liquid-cooling assembly and power supply. The components separate into an organized layered arrangement around the case while retaining their original shapes. The case base stays on the desk. The camera slowly orbits to the right around the computer from the starting three-quarter view. The blue-white computer lighting and surrounding room remain consistent throughout the continuous shot."
       ],
       "color": "#232",
       "bgcolor": "#353"
@@ -874,8 +912,8 @@
         3390
       ],
       "size": [
-        580,
-        350
+        620,
+        390
       ],
       "flags": {
         "collapsed": false
@@ -885,55 +923,55 @@
       "inputs": [
         {
           "label": "audio_vae",
-          "name": "ckpt_name",
+          "name": "audio_vae",
           "type": "COMBO",
           "widget": {
-            "name": "ckpt_name"
+            "name": "audio_vae"
           },
           "link": null
         },
         {
           "label": "video_vae",
-          "name": "gemma_path",
+          "name": "video_vae",
           "type": "COMBO",
           "widget": {
-            "name": "gemma_path"
+            "name": "video_vae"
           },
           "link": null
         },
         {
           "label": "ltx_2.5_model_checkpoint",
-          "name": "gemma_path_1",
+          "name": "ltx_2.5_model_checkpoint",
           "type": "COMBO",
           "widget": {
-            "name": "gemma_path_1"
+            "name": "ltx_2.5_model_checkpoint"
           },
           "link": null
         },
         {
-          "label": "spacial_latent_upscaler_model",
-          "name": "clip_name",
+          "label": "spatial_latent_upscaler_model",
+          "name": "spatial_latent_upscaler_model",
           "type": "COMBO",
           "widget": {
-            "name": "clip_name"
+            "name": "spatial_latent_upscaler_model"
           },
           "link": null
         },
         {
           "label": "gemma4_e2b_enhancer",
-          "name": "clip_name_1",
+          "name": "gemma4_e2b_enhancer",
           "type": "COMBO",
           "widget": {
-            "name": "clip_name_1"
+            "name": "gemma4_e2b_enhancer"
           },
           "link": null
         },
         {
           "label": "gemma4_12b_encoder",
-          "name": "model_name",
+          "name": "gemma4_12b_encoder",
           "type": "COMBO",
           "widget": {
-            "name": "model_name"
+            "name": "gemma4_12b_encoder"
           },
           "link": null
         }
@@ -1010,7 +1048,15 @@
         "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
         "gemma4_e2b_it_bf16.safetensors",
         "gemma4-12b-with-proj-ltx-2.5-bf16.safetensors"
-      ]
+      ],
+      "widgets_values_named": {
+        "audio_vae": "ltx-2.5-audio-vae-bf16.safetensors",
+        "video_vae": "ltx-2.5-video-vae-bf16.safetensors",
+        "ltx_2.5_model_checkpoint": "ltx-2.5-22b-distilled-transformer-bf16.safetensors",
+        "spatial_latent_upscaler_model": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+        "gemma4_e2b_enhancer": "gemma4_e2b_it_bf16.safetensors",
+        "gemma4_12b_encoder": "gemma4-12b-with-proj-ltx-2.5-bf16.safetensors"
+      }
     },
     {
       "id": 5517,
@@ -1123,14 +1169,21 @@
         "euler",
         1,
         42
-      ]
+      ],
+      "widgets_values_named": {
+        "bypass": false,
+        "strength": 1,
+        "sampler_name": "euler",
+        "cfg": 1,
+        "noise_seed": 42
+      }
     },
     {
       "id": 5512,
       "type": "PrimitiveFloat",
       "pos": [
         -3570,
-        4060
+        4170
       ],
       "size": [
         350,
@@ -1218,13 +1271,40 @@
       "mode": 0,
       "inputs": [],
       "outputs": [],
-      "title": "XPLDV — BASIC, COMPLEX and settings",
+      "title": "XPLDV - settings and COMPLEX template",
       "properties": {},
       "widgets_values": [
-        "# XPLDV — prompts and settings\n\nStart with LoRA strength **0.5**. The connected LoRA feeds both sampling stages.\n\n| Setting | Value |\n| --- | --- |\n| Duration | 10 seconds; 241 frames at 24 FPS |\n| First stage | 960 × 544; Euler ancestral; 8 steps |\n| Second stage | 1920 × 1088; Euler; 3 steps |\n| Image conditioning | 0.7 in stage 1; 1.0 in stage 2 |\n| CFG | 1 in both stages |\n| Seed for comparisons | 42 in both stages |\n| Second-stage manual sigmas | `0.85, 0.7250, 0.4219, 0.0` |\n| Prompt enhancement | Off |\n\nKeep the first-stage schedule unchanged. The second-stage sigmas belong in **Upscale and re-sampler (3 steps)**.\n\n## BASIC — a short reveal\n\n```text\n[SUBJECT] separates into an ordered exploded view. [NAMED OUTER PARTS] move\noutward along controlled paths, revealing [NAMED INTERNAL COMPONENTS]\nunderneath. The parts settle into an organized [LAYERED / RADIAL] arrangement\naround [EXPOSED CORE] while retaining their original shapes. [CAMERA\nBEHAVIOR]. [LIGHTING CONSISTENT WITH THE IMAGE] remains consistent throughout\nthe continuous shot.\n```\n\n## COMPLEX — a detailed, cinematic reveal\n\nA COMPLEX prompt guides a more detailed, cinematic reveal through successive layers, describing deeper internal components, their appearance, and their positions, including details hidden or difficult to see in the starting image. These descriptions give LTX 2.5 the information needed to construct the intended reveal beyond what the image alone shows.\n\n```text\n[SUBJECT] separates into an ordered exploded view while [CONTINUING ACTION OR\nANCHORED SUPPORT]. [OUTER PART A] detaches and translates [DIRECTION], keeping\n[ORIENTATION]. [OUTER PART B] moves [DISTINCT PATH], and [OTHER OUTER PARTS]\nmove [PATHS]. These exterior parts move first and travel furthest, leaving\nclear gaps. They reveal [SPECIFIC COMPONENT AND RECOGNIZABLE FEATURES] at\n[POSITION RELATIVE TO THE OBJECT], with [OTHER COMPONENTS AND THEIR\nLOCATIONS]. [CORE COMPONENTS] remain mounted together. Each detached section\nretains its shape as the arrangement settles and holds steady. [CAMERA\nBEHAVIOR]. [CONSISTENT LIGHTING] continues through one continuous shot.\n```\n\nFor a COMPLEX prompt, privately plan five major assemblies or reveal stages from the outer parts toward the deeper structure. Follow the object’s construction; the plan does not require exactly five detached pieces.\n\nKeep related parts within a named internal mechanism intact when useful, such as the gears inside a watch movement. This does not keep the exterior attached. Describe which internal layers also move; omit the sentence about components remaining mounted together when it does not fit the reveal.\n\nReplace the brackets with the subject and its parts; omit clauses that do not apply to the intended shot. Describe the final arrangement as a vertical stack, horizontal row, or radial arrangement around the center. Use “layered” only for a stack. Describe simultaneous movement or a clear sequence without mixing the two. Keep a moving subject and its separated parts traveling together, and leave room in the frame for the reveal.\n\nRecommended prompt phrase: **\"separates into an ordered exploded view.\"** This wording appears in all 204 separation training captions. “Separates into individual components” is another phrasing we tested. With either opening, describe the movement and the final arrangement. Optional sound effects, ambience, music, or dialogue can go in a final sentence. Audio wording can influence the action; keep the audio pipeline connected.\n\n## LoRA strength\n\n| Strength | Use |\n| --- | --- |\n| 0.5 | Recommended starting point. |\n| 0.75 | Stronger influence; may improve separation but can simplify detail or repeat parts. |\n\nChange one setting at a time while keeping the image, prompt, and seed fixed.\n"
+        "# XPLDV — prompts and settings\n\nStart with LoRA strength **0.5**. The connected LoRA feeds both sampling stages.\n\n| Setting | Value |\n| --- | --- |\n| Duration | 10 seconds; 241 frames at 24 FPS |\n| First stage | 960 × 544; Euler ancestral; 8 steps |\n| Second stage | 1920 × 1088; Euler; 3 steps |\n| Image conditioning | 0.7 in stage 1; 1.0 in stage 2 |\n| CFG | 1 in both stages |\n| Seed for comparisons | 42 in both stages |\n| Second-stage manual sigmas | `0.85, 0.7250, 0.4219, 0.0` |\n| Prompt enhancement | Can be enabled; off in the recorded example |\n\nEnabling **Enhance positive prompt** works with XPLDV, but results may vary. It has not been extensively tested. The recorded example uses enhancement off to preserve its supplied prompt.\n\nKeep the first-stage schedule unchanged. The second-stage sigmas belong in **Upscale and re-sampler (3 steps)**.\n\nThe reusable BASIC template is in the note directly above the positive prompt. The prompt is prefilled with the tested desktop-PC example.\n\n## COMPLEX — a detailed, cinematic reveal\n\nA COMPLEX prompt guides a more detailed, cinematic reveal through successive layers, describing deeper internal components, their appearance, and their positions, including details hidden or difficult to see in the starting image. These descriptions give LTX 2.5 the information needed to construct the intended reveal beyond what the image alone shows.\n\n```text\n[SUBJECT] separates into an ordered exploded view while [CONTINUING ACTION OR ANCHORED SUPPORT]. [OUTER PART A] detaches and translates [DIRECTION], keeping [ORIENTATION]. [OUTER PART B] moves [DISTINCT PATH], and [OTHER OUTER PARTS] move [PATHS]. These exterior parts move first and travel furthest, leaving clear gaps. They reveal [SPECIFIC COMPONENT AND RECOGNIZABLE FEATURES] at [POSITION RELATIVE TO THE OBJECT], with [OTHER COMPONENTS AND THEIR LOCATIONS]. [CORE COMPONENTS] remain mounted together. Each detached section retains its shape as the arrangement settles and holds steady. [CAMERA BEHAVIOR]. [CONSISTENT LIGHTING] continues through one continuous shot.\n```\n\nFor a COMPLEX prompt, privately plan five major assemblies or reveal stages from the outer parts toward the deeper structure. Follow the object’s construction; the plan does not require exactly five detached pieces.\n\nKeep related parts within a named internal mechanism intact when useful, such as the gears inside a watch movement. This does not keep the exterior attached. Describe which internal layers also move; omit the sentence about components remaining mounted together when it does not fit the reveal.\n\nReplace the brackets with the subject and its parts; omit clauses that do not apply to the intended shot. Describe the final arrangement as a vertical stack, horizontal row, or radial arrangement around the center. Use “layered” only for a stack. Describe simultaneous movement or a clear sequence without mixing the two. Keep a moving subject and its separated parts traveling together, and leave room in the frame for the reveal.\n\nRecommended prompt phrase: **\"separates into an ordered exploded view.\"** This wording appears in all 204 separation training captions. “Separates into individual components” is another phrasing we tested. With either opening, describe the movement and the final arrangement. Optional sound effects, ambience, music, or dialogue can go in a final sentence. Audio wording can influence the action; keep the audio pipeline connected.\n\n## LoRA strength\n\n| Strength | Use |\n| --- | --- |\n| 0.5 | Recommended starting point. |\n| 0.75 | Stronger influence; may improve separation but can simplify detail or repeat parts. |\n\nChange one setting at a time while keeping the image, prompt, and seed fixed.\n"
       ],
       "color": "#322",
       "bgcolor": "#533"
+    },
+    {
+      "id": 5580,
+      "type": "MarkdownNote",
+      "pos": [
+        -3570,
+        2800
+      ],
+      "size": [
+        350,
+        350
+      ],
+      "flags": {},
+      "order": 22,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [],
+      "title": "Reusable BASIC template",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "Node name for S&R": "MarkdownNote"
+      },
+      "widgets_values": [
+        "## Reusable BASIC template\n\nCopy this template into the positive prompt below when using your own image, then replace the bracketed fields.\n\n[SUBJECT] separates into an ordered exploded view. [NAMED OUTER PARTS] move outward along controlled paths, revealing [NAMED INTERNAL COMPONENTS] underneath. The parts settle into an organized [LAYERED / RADIAL] arrangement around [EXPOSED CORE] while retaining their original shapes. [CAMERA BEHAVIOR]. [LIGHTING CONSISTENT WITH THE IMAGE] remains consistent throughout the continuous shot."
+      ],
+      "color": "#243a35",
+      "bgcolor": "#344b43"
     }
   ],
   "links": [
@@ -2023,7 +2103,7 @@
               "Node name for S&R": "PrimitiveBoolean"
             },
             "widgets_values": [
-              false
+              true
             ]
           },
           {
@@ -2114,7 +2194,7 @@
               1,
               0,
               1.15,
-              0,
+              457289196,
               0,
               false,
               true
@@ -2208,7 +2288,7 @@
               1,
               0,
               1.15,
-              0,
+              457289196,
               0,
               false,
               true
@@ -2702,7 +2782,7 @@
               "Node name for S&R": "PrimitiveFloat"
             },
             "widgets_values": [
-              5
+              10
             ],
             "color": "#2a363b",
             "bgcolor": "#3f5159"
@@ -2897,8 +2977,8 @@
             "widgets_values": [
               "",
               "",
-              "ltx-2.3-22b-dev.safetensors",
-              "ltx-2.3-22b-dev.safetensors"
+              false,
+              "ltx-2.5-22b-distilled-transformer-bf16.safetensors"
             ]
           },
           {
@@ -2964,7 +3044,7 @@
               "",
               "",
               false,
-              "ltx-2.3-22b-dev.safetensors"
+              "ltx-2.5-22b-distilled-transformer-bf16.safetensors"
             ]
           }
         ],
@@ -4677,7 +4757,7 @@
               "Node name for S&R": "RandomNoise"
             },
             "widgets_values": [
-              43,
+              42,
               "fixed"
             ]
           },
@@ -5489,8 +5569,8 @@
             "widgets_values": [
               512,
               64,
-              64,
-              8
+              128,
+              32
             ]
           }
         ],
@@ -6142,7 +6222,7 @@
               "Node name for S&R": "KSamplerSelect"
             },
             "widgets_values": [
-              "euler_ancestral"
+              "euler"
             ]
           },
           {
@@ -6727,7 +6807,7 @@
           "lastLinkId": 13841,
           "lastRerouteId": 5036
         },
-        "revision": 0,
+        "revision": 1,
         "config": {},
         "name": "Load Models",
         "inputNode": {
@@ -6751,7 +6831,7 @@
         "inputs": [
           {
             "id": "1ba7f82e-b1b9-49ba-b336-309da4444c01",
-            "name": "ckpt_name",
+            "name": "audio_vae",
             "type": "COMBO",
             "linkIds": [
               13797
@@ -6764,7 +6844,7 @@
           },
           {
             "id": "d54e81a5-0f60-4bc2-abbf-2c8bbeb33d15",
-            "name": "gemma_path",
+            "name": "video_vae",
             "type": "COMBO",
             "linkIds": [
               13799
@@ -6777,7 +6857,7 @@
           },
           {
             "id": "50ae2205-f683-47cb-b812-fe649d3b9bb6",
-            "name": "gemma_path_1",
+            "name": "ltx_2.5_model_checkpoint",
             "type": "COMBO",
             "linkIds": [
               13801,
@@ -6791,12 +6871,12 @@
           },
           {
             "id": "fd473638-f7cf-452f-9cd7-933f95a5884a",
-            "name": "clip_name",
+            "name": "spatial_latent_upscaler_model",
             "type": "COMBO",
             "linkIds": [
               13810
             ],
-            "label": "spacial_latent_upscaler_model",
+            "label": "spatial_latent_upscaler_model",
             "pos": [
               -5237.3611236420875,
               3397.364324949378
@@ -6804,7 +6884,7 @@
           },
           {
             "id": "9fae03c8-4de7-44b5-9710-f3feb741c769",
-            "name": "clip_name_1",
+            "name": "gemma4_e2b_enhancer",
             "type": "COMBO",
             "linkIds": [
               13811
@@ -6817,7 +6897,7 @@
           },
           {
             "id": "f1d91e85-1ca3-4096-a742-24747f61ae4c",
-            "name": "model_name",
+            "name": "gemma4_12b_encoder",
             "type": "COMBO",
             "linkIds": [
               13812
@@ -6961,7 +7041,7 @@
                 ]
               }
             ],
-            "title": "Load VAE - Audio",
+            "title": "Audio VAE + vocoder",
             "properties": {
               "cnr_id": "comfy-core",
               "ver": "0.32.0",
@@ -7006,14 +7086,14 @@
                 ]
               }
             ],
-            "title": "Load VAE - Video",
+            "title": "Video diffusion VAE",
             "properties": {
               "cnr_id": "comfy-core",
               "ver": "0.32.0",
               "Node name for S&R": "VAELoader"
             },
             "widgets_values": [
-              "ltx-2.5-audio-vae-bf16.safetensors"
+              "ltx-2.5-video-vae-bf16.safetensors"
             ]
           },
           {
@@ -7057,9 +7137,10 @@
               "Node name for S&R": "UNETLoader"
             },
             "widgets_values": [
-              "ltx-2.5-22b-dev-transformer-bf16.safetensors",
+              "ltx-2.5-22b-distilled-transformer-bf16.safetensors",
               "default"
-            ]
+            ],
+            "title": "LTX 2.5 distilled BF16"
           },
           {
             "id": 5573,
@@ -7139,8 +7220,9 @@
               "Node name for S&R": "LatentUpscaleModelLoader"
             },
             "widgets_values": [
-              "ltx-2.3-spatial-upscaler-x2-1.1.safetensors"
-            ]
+              "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+            ],
+            "title": "LTX 2.5 spatial upscaler 2x"
           },
           {
             "id": 5572,
@@ -7177,14 +7259,14 @@
                 ]
               }
             ],
-            "title": "Load CLIP - Text Enhancer",
+            "title": "Gemma 4 E2B - prompt enhancer",
             "properties": {
               "cnr_id": "comfy-core",
               "ver": "0.32.0",
               "Node name for S&R": "CLIPLoader"
             },
             "widgets_values": [
-              "ViT-B-32.pt",
+              "gemma4_e2b_it_bf16.safetensors",
               "ltxv",
               "default"
             ]
@@ -7224,14 +7306,14 @@
                 ]
               }
             ],
-            "title": "Load CLIP - Text Encoder",
+            "title": "Gemma 4 12B - text encoder",
             "properties": {
               "cnr_id": "comfy-core",
               "ver": "0.32.0",
               "Node name for S&R": "CLIPLoader"
             },
             "widgets_values": [
-              "ViT-B-32.pt",
+              "gemma4-12b-with-proj-ltx-2.5-bf16.safetensors",
               "ltxv",
               "default"
             ]
@@ -7877,4 +7959,4 @@
     ]
   },
   "version": 0.4
-}
+}