Skip to content
Python

26 apps más de plantilla: video, dos referencias y LoRAs de cámara

neuregexcd930ece786d
diff --git a/README.md b/README.md
index 9a254b9..2b5df3a 100644
--- a/README.md
+++ b/README.md
@@ -17,9 +17,9 @@ browser ──▶ web/ + server/ (Node, Comfy SDK) ──▶ gateway :8000 ─
| | |
|---|---|
-| Apps | 70 manifests, each with a workflow: 29 *ready* that ran on a GPU, 40 template apps (a fixed prompt over the `edit` or `z-image` pipeline, written by `tools/template_apps.py`; their lab run is pending) and 1 *draft* (`reshoot`: it runs, but parts of the new view come out unfilled) |
+| Apps | 96 manifests, each with a workflow: 29 *ready* that ran on a GPU, 66 template apps (a fixed prompt over the `edit`, `z-image`, `video` or `relight` pipeline, written by `tools/template_apps.py`; their lab runs are pending) and 1 *draft* (`reshoot`: it runs, but parts of the new view come out unfilled) |
| Workflows | Built from official sources by `tools/build_apps.py` (the two smoke apps are hand-written) and accepted by ComfyUI's validator: ComfyUI `e3bee7f` (0.42.0), ComfyUI-LTXVideo `44c8a36`, ComfyUI-CrossViewWarp `fbb1e93` |
-| LoRA lab | 141 cases, one showcase per app plus one per effect (35); the prompt each one would queue (values, overrides and inserted LoRAs) passes the same validator |
+| LoRA lab | 168 cases, one showcase per app plus one per effect (35); the prompt each one would queue (values, overrides and inserted LoRAs) passes the same validator |
| End to end | Web UI → server → gateway → comfy-api-proxy → ComfyUI tested with the model-free smoke apps; lab chaining tested the same way |
| CI | `.github/workflows/validate.yml` runs all of the above on every push, including a CPU-only ComfyUI that checks every workflow and lab case (about 3 minutes) |
| GPU runs | Showcase lab run on an L40S (`20261020-133620-l40s`): 35 of 36 cases, one example per app; `recast` failed there and passed in a re-run (`20261020-162214-l40s`). Effects on 4 H100s at once (`20261021-140940-h100`, `--shards 4`): 61 of 61 cases in about 31 minutes of GPU, 2 min 52 s per effect on average, peak 30.8 GB. Two more H100 rounds (`20261021-152107-h100`, `20261021-175735-h100`) re-ran the 5 that failed with new photos and tried 7 LoRAs by other authors: Painting, Pirate captain, Zen and Crash zoom out were taken out, Dolly zoom and Bullet time moved to ostris's and joyfox's LoRAs, and Money rain, Clones and Steam from nose and ears were added. A fourth round (`20261021-195731-h100`, 14 of 14 cases) ran the new ones as shipped; after it, Jungle, Samurai, Warrior, Jumpscare, Smoke outfit change and Gothic transformation were taken out to keep the app SFW. That leaves 35 effects, each with an example in the gallery. VRAM figures below are weight sizes; the lab's measured peaks show up in the web UI |
@@ -56,11 +56,14 @@ browser ──▶ web/ + server/ (Node, Comfy SDK) ──▶ gateway :8000 ─
### Template apps
-Higgsfield's catalog is mostly fixed prompts over a few pipelines, and so is this: each of these apps is the `edit`
-pipeline (FLUX 3 Image Klein, one reference image) or Z-Image Turbo with its prompt baked in, selects that fill parts of
-it, and an "Extra direction" box appended to it. `tools/template_apps.py` holds the table and writes the 40 folders
-(manifest + a copy of the base workflow) and their lab cases; `--readme` prints this list and `--lab` the command that
-makes every example. Add a line to the table there, not a folder by hand.
+Higgsfield's catalog is mostly fixed prompts over a few pipelines, and so is this: each of these apps is one of
+Freefield's pipelines with its prompt baked in, selects that fill parts of it, and an "Extra direction" box appended
+to it. The bases: `edit` (FLUX 3 Image Klein, one reference image), `edit` with a second reference (a garment, another
+person, a face), `z-image` (text to image), `video` (LTX-2.5 image-to-video with its own audio) and `relight`'s
+Qwen-Image-Edit-2509 pipeline with another LoRA (dx8152's Multiple Angles for `angles`, `shots` and `zooms`; lovis93's
+next-scene v2 for `whats-next`). `tools/template_apps.py` holds the table and writes the 66 folders (manifest + the
+base workflow, with the few edits a base needs) and their lab cases; `--readme` prints this list and `--lab` the
+command that makes every example. Add a line to the table there, not a folder by hand.
| App | Rebuilds | Base | Inputs |
|---|---|---|---|
@@ -104,6 +107,32 @@ makes every example. Add a line to the table there, not a folder by hand.
| `edit-text` | Edit Text | `edit` | image with text, text to replace, new text |
| `image-effects` | Image Effects | `edit` | your photo, effect |
| `color-grade` | Color Grading | `edit` | your photo, look |
+| `fashion-factory` | Fashion Factory | `edit` | your photo, garment photo, scene |
+| `try-on` | AI Stylist | `edit` | your photo, garment photo |
+| `outfit-swap` | Outfit Swap | `edit` | your photo, photo of the outfit |
+| `outfit-shot` | Outfit Shot | `edit` | model photo, garment photo, pose |
+| `character-swap` | Character Swap | `edit` | scene photo, photo of the new person |
+| `face-swap` | Face Swap | `edit` | photo to edit, photo of the face |
+| `behind-the-scenes` | Behind the Scenes | `video` | your photo |
+| `asmr-product` | ASMR Add-On | `video` | product photo |
+| `asmr-whisper` | ASMR Classic | `video` | your photo |
+| `asmr-host` | ASMR Host | `video` | your photo |
+| `chameleon` | Chameleon | `video` | product photo |
+| `magic-button` | Magic button | `video` | product photo |
+| `truck-ad` | Truck Ad | `video` | product photo |
+| `kick-ad` | Kick Ad | `video` | product photo |
+| `volcano-ad` | Volcano Ad | `video` | product photo |
+| `roller-coaster` | Roller Coaster | `video` | your photo |
+| `storm-creature` | Storm Creature | `video` | your photo |
+| `cloud-surf` | Cloud Surf | `video` | your photo |
+| `mukbang` | Mukbang | `video` | your photo |
+| `melting-doodle` | Melting Doodle | `video` | your photo |
+| `rap-video` | Rap God | `video` | your photo |
+| `idol` | Idol | `video` | your photo |
+| `angles` | Angles 2.0 | `relight + Qwen-Edit-2509-Multiple-angles` | your photo, camera move |
+| `shots` | Shots | `relight + Qwen-Edit-2509-Multiple-angles` | your photo, shot |
+| `zooms` | Zooms | `relight + Qwen-Edit-2509-Multiple-angles` | your photo, zoom |
+| `whats-next` | What's Next? | `relight + next-scene-qwen-image-lora-2509` | your photo, continuation |
Full table: [`docs/vram-map.md`](docs/vram-map.md) (`python tools/catalog.py vram`). Between two apps with different
base models the gateway calls ComfyUI's `POST /free`, so nothing from the previous app stays in VRAM.
diff --git a/apps/angles/app.json b/apps/angles/app.json
new file mode 100644
index 0000000..f830907
--- /dev/null
+++ b/apps/angles/app.json
@@ -0,0 +1,166 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "angles",
+ "name": "Angles",
+ "section": "image",
+ "summary": "The same scene from another camera position: move, rotate, look down (Multiple Angles LoRA on Qwen-Image-Edit).",
+ "higgsfield_feature": "Angles 2.0",
+ "pack": "image-qwen-edit",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_qwen_image_edit_2509_relight.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/relight's (built by tools/build_apps.py) with 镜头转换.safetensors in place of the Relight LoRA and no Relight trigger; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "move",
+ "type": "select",
+ "label": "Camera move",
+ "default": "将镜头向左旋转45度 Rotate the camera 45 degrees to the left.",
+ "options": [
+ {
+ "value": "将镜头向前移动 Move the camera forward.",
+ "label": "Move forward"
+ },
+ {
+ "value": "将镜头向后移动 Move the camera backward.",
+ "label": "Move back"
+ },
+ {
+ "value": "将镜头向左移动 Move the camera left.",
+ "label": "Move left"
+ },
+ {
+ "value": "将镜头向右移动 Move the camera right.",
+ "label": "Move right"
+ },
+ {
+ "value": "将镜头向上移动 Move the camera up.",
+ "label": "Move up"
+ },
+ {
+ "value": "将镜头向下移动 Move the camera down.",
+ "label": "Move down"
+ },
+ {
+ "value": "将镜头向左旋转45度 Rotate the camera 45 degrees to the left.",
+ "label": "Rotate 45° left"
+ },
+ {
+ "value": "将镜头向右旋转45度 Rotate the camera 45 degrees to the right.",
+ "label": "Rotate 45° right"
+ },
+ {
+ "value": "将镜头转为俯视 Turn the camera to a top-down view.",
+ "label": "Top-down view"
+ },
+ {
+ "value": "将镜头转为仰视 Turn the camera to a low-angle view looking up.",
+ "label": "Low angle"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "15",
+ "field": "value",
+ "template": "{move} {details}"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "10:3",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "12",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Comfy-Org/Qwen-Image-Edit_ComfyUI",
+ "file": "split_files/diffusion_models/qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "dir": "diffusion_models",
+ "gb": 20.43,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "Comfy-Org/Qwen-Image_ComfyUI",
+ "file": "split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors",
+ "dir": "text_encoders",
+ "gb": 9.38,
+ "phase": "encode",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "Comfy-Org/Qwen-Image_ComfyUI",
+ "file": "split_files/vae/qwen_image_vae.safetensors",
+ "dir": "vae",
+ "gb": 0.25,
+ "phase": "decode",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "lightx2v/Qwen-Image-Lightning",
+ "file": "Qwen-Image-Edit-2509/Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors",
+ "dir": "loras",
+ "gb": 0.85,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "dx8152/Qwen-Edit-2509-Multiple-angles",
+ "file": "镜头转换.safetensors",
+ "dir": "loras",
+ "gb": 0.24,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ }
+ ],
+ "credits": [
+ "Qwen-Image-Edit-2509 by Alibaba Qwen (Apache-2.0)",
+ "Multiple Angles LoRA by dx8152 (Apache-2.0)",
+ "Lightning LoRA by lightx2v",
+ "Workflow: ComfyUI official template"
+ ],
+ "test": {
+ "image": "test/portrait.png"
+ }
+}
diff --git a/apps/angles/workflow.api.json b/apps/angles/workflow.api.json
new file mode 100644
index 0000000..2c8e443
--- /dev/null
+++ b/apps/angles/workflow.api.json
@@ -0,0 +1,251 @@
+{
+ "10:3": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "10:75",
+ 0
+ ],
+ "positive": [
+ "10:111",
+ 0
+ ],
+ "negative": [
+ "10:110",
+ 0
+ ],
+ "latent_image": [
+ "10:88",
+ 0
+ ],
+ "seed": 1118877715456453,
+ "steps": 4,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "10:8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "10:3",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "10:37": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "10:38": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen_2.5_vl_7b_fp8_scaled.safetensors",
+ "type": "qwen_image",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "10:39": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "qwen_image_vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "10:66": {
+ "class_type": "ModelSamplingAuraFlow",
+ "inputs": {
+ "model": [
+ "10:118",
+ 0
+ ],
+ "shift": 3
+ },
+ "_meta": {
+ "title": "ModelSamplingAuraFlow"
+ }
+ },
+ "10:75": {
+ "class_type": "CFGNorm",
+ "inputs": {
+ "model": [
+ "10:66",
+ 0
+ ],
+ "strength": 1,
+ "pre_cfg": false
+ },
+ "_meta": {
+ "title": "CFGNorm"
+ }
+ },
+ "10:88": {
+ "class_type": "VAEEncode",
+ "inputs": {
+ "pixels": [
+ "10:117",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Encode"
+ }
+ },
+ "10:89": {
+ "class_type": "LoraLoaderModelOnly",
+ "inputs": {
+ "model": [
+ "10:37",
+ 0
+ ],
+ "lora_name": "Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors",
+ "strength_model": 1
+ },
+ "_meta": {
+ "title": "Load LoRA"
+ }
+ },
+ "10:110": {
+ "class_type": "TextEncodeQwenImageEditPlus",
+ "inputs": {
+ "clip": [
+ "10:38",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ],
+ "image1": [
+ "10:117",
+ 0
+ ],
+ "prompt": ""
+ },
+ "_meta": {
+ "title": "Text Encode Qwen Image Edit Plus"
+ }
+ },
+ "10:111": {
+ "class_type": "TextEncodeQwenImageEditPlus",
+ "inputs": {
+ "clip": [
+ "10:38",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ],
+ "image1": [
+ "10:117",
+ 0
+ ],
+ "prompt": [
+ "14",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Text Encode Qwen Image Edit Plus"
+ }
+ },
+ "10:117": {
+ "class_type": "FluxKontextImageScale",
+ "inputs": {
+ "image": [
+ "11",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "FluxKontextImageScale"
+ }
+ },
+ "10:118": {
+ "class_type": "LoraLoaderModelOnly",
+ "inputs": {
+ "model": [
+ "10:89",
+ 0
+ ],
+ "lora_name": "镜头转换.safetensors",
+ "strength_model": 1
+ },
+ "_meta": {
+ "title": "Load LoRA"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "warm_living_room.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "12": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "10:8",
+ 0
+ ],
+ "filename_prefix": "freefield/angles"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "14": {
+ "class_type": "StringConcatenate",
+ "inputs": {
+ "string_b": [
+ "15",
+ 0
+ ],
+ "string_a": "",
+ "delimiter": ""
+ },
+ "_meta": {
+ "title": "LoRA trigger word"
+ }
+ },
+ "15": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": " relight the image using twilight light (with a warm color temperature, soft light, and blurred shadows)"
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ }
+}
diff --git a/apps/asmr-host/app.json b/apps/asmr-host/app.json
new file mode 100644
index 0000000..65b94ed
--- /dev/null
+++ b/apps/asmr-host/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "asmr-host",
+ "name": "ASMR Host",
+ "section": "video",
+ "summary": "You hosting an ASMR studio with a big microphone and fairy lights.",
+ "higgsfield_feature": "ASMR Host",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "The person sits in a cozy ASMR studio with a large microphone in front of them, fairy lights behind, and whispers into the mic while gently handling a soft brush; close whispering sound. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "9:16 (Portrait Widescreen)"
+ }
+}
diff --git a/apps/asmr-host/workflow.api.json b/apps/asmr-host/workflow.api.json
new file mode 100644
index 0000000..4955576
--- /dev/null
+++ b/apps/asmr-host/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/asmr-host",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/asmr-product/app.json b/apps/asmr-product/app.json
new file mode 100644
index 0000000..78dce07
--- /dev/null
+++ b/apps/asmr-product/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "asmr-product",
+ "name": "ASMR Product",
+ "section": "video",
+ "summary": "Your product in an ASMR clip: tapping, scratching, no music.",
+ "higgsfield_feature": "ASMR Add-On",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Product photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "An ASMR video: hands with manicured nails slowly tap and scratch the product, turning it under soft light, extreme close-up; crisp tapping and scratching sounds, no music, no voice. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "9:16 (Portrait Widescreen)"
+ }
+}
diff --git a/apps/asmr-product/workflow.api.json b/apps/asmr-product/workflow.api.json
new file mode 100644
index 0000000..ce95f13
--- /dev/null
+++ b/apps/asmr-product/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/asmr-product",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/asmr-whisper/app.json b/apps/asmr-whisper/app.json
new file mode 100644
index 0000000..1c85834
--- /dev/null
+++ b/apps/asmr-whisper/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "asmr-whisper",
+ "name": "ASMR Whisper",
+ "section": "video",
+ "summary": "You whispering and tapping in a classic ASMR clip.",
+ "higgsfield_feature": "ASMR Classic",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "An ASMR video: the person whispers softly toward the camera and taps their fingers on the table, slow gentle movements, warm low light; soft whispering and tapping sounds, no music. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "9:16 (Portrait Widescreen)"
+ }
+}
diff --git a/apps/asmr-whisper/workflow.api.json b/apps/asmr-whisper/workflow.api.json
new file mode 100644
index 0000000..41841f3
--- /dev/null
+++ b/apps/asmr-whisper/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/asmr-whisper",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/behind-the-scenes/app.json b/apps/behind-the-scenes/app.json
new file mode 100644
index 0000000..b28b6c4
--- /dev/null
+++ b/apps/behind-the-scenes/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "behind-the-scenes",
+ "name": "Behind the Scenes",
+ "section": "video",
+ "summary": "The camera pulls back from your photo to reveal the film set around it.",
+ "higgsfield_feature": "Behind the Scenes",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "The camera pulls back from the photo to reveal a film set around it: crew members, a boom microphone, softbox lights and a clapperboard, people moving about; the sound of a busy set, a director calling \"rolling\". {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "16:9 (Widescreen)"
+ }
+}
diff --git a/apps/behind-the-scenes/workflow.api.json b/apps/behind-the-scenes/workflow.api.json
new file mode 100644
index 0000000..d62ba71
--- /dev/null
+++ b/apps/behind-the-scenes/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/behind-the-scenes",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/chameleon/app.json b/apps/chameleon/app.json
new file mode 100644
index 0000000..9e387c8
--- /dev/null
+++ b/apps/chameleon/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "chameleon",
+ "name": "Chameleon",
+ "section": "video",
+ "summary": "A chameleon climbs onto your product and takes its colors.",
+ "higgsfield_feature": "Chameleon",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Product photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "A chameleon crawls onto the product and slowly changes its skin to the exact colors and pattern of the product, blinking, macro lens; soft ambient jungle sound. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "16:9 (Widescreen)"
+ }
+}
diff --git a/apps/chameleon/workflow.api.json b/apps/chameleon/workflow.api.json
new file mode 100644
index 0000000..7d862e7
--- /dev/null
+++ b/apps/chameleon/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/chameleon",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/character-swap/app.json b/apps/character-swap/app.json
new file mode 100644
index 0000000..f40ba57
--- /dev/null
+++ b/apps/character-swap/app.json
@@ -0,0 +1,133 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "character-swap",
+ "name": "Character Swap",
+ "section": "image",
+ "summary": "Replaces the person in a scene with the person from a second photo.",
+ "higgsfield_feature": "Character Swap",
+ "pack": "image-flux3",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_flux3_klein_distilled_image_edit.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/edit's (built by tools/build_apps.py) plus a second LoadImage on ReferenceMultiLatent's images.image1; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Scene photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "image2",
+ "type": "image",
+ "label": "Photo of the new person",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "12",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "4",
+ "field": "text",
+ "template": "The scene from <ref_image_0> with its main person replaced by the person from <ref_image_1>: same pose, framing, lighting and clothing as the scene; the new person's face, hair and build. {details}"
+ }
+ ]
+ },
+ {
+ "id": "steps",
+ "type": "slider",
+ "label": "Steps",
+ "min": 8,
+ "max": 32,
+ "step": 1,
+ "default": 16,
+ "bind": [
+ {
+ "node": "7",
+ "field": "steps"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "7",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "9",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "diffusion_models/flux3-image-klein-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 5.83,
+ "phase": "denoise",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "text_encoders/qwen3vl_4b_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 4.86,
+ "phase": "encode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "vae/flux2-vae.safetensors",
+ "dir": "vae",
+ "gb": 0.34,
+ "phase": "decode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ }
+ ],
+ "credits": [
+ "FLUX 3 Image Klein by Black Forest Labs (FLUX Kommunity License)"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "image2": "test/second.png"
+ }
+}
diff --git a/apps/character-swap/workflow.api.json b/apps/character-swap/workflow.api.json
new file mode 100644
index 0000000..401c354
--- /dev/null
+++ b/apps/character-swap/workflow.api.json
@@ -0,0 +1,164 @@
+{
+ "1": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "flux3-image-klein-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "2": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen3vl_4b_int8_convrot.safetensors",
+ "type": "flux3",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "3": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "flux2-vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "4": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "2",
+ 0
+ ],
+ "text": "Change the jacket to deep red leather, keep everything else the same."
+ },
+ "_meta": {
+ "title": "Positive Prompt"
+ }
+ },
+ "7": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "1",
+ 0
+ ],
+ "positive": [
+ "17",
+ 0
+ ],
+ "negative": [
+ "17",
+ 1
+ ],
+ "latent_image": [
+ "17",
+ 2
+ ],
+ "seed": 563228241303057,
+ "steps": 16,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "7",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "9": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "8",
+ 0
+ ],
+ "filename_prefix": "freefield/character-swap"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "woman_gold_earring_teal_top_portrait.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "17": {
+ "class_type": "ReferenceMultiLatent",
+ "inputs": {
+ "positive": [
+ "4",
+ 0
+ ],
+ "negative": [
+ "32",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ],
+ "images.image0": [
+ "11",
+ 0
+ ],
+ "resolution": "resize",
+ "resolution.megapixels": 1,
+ "images.image1": [
+ "12",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Set Reference Multi Latent"
+ }
+ },
+ "32": {
+ "class_type": "ConditioningZeroOut",
+ "inputs": {
+ "conditioning": [
+ "4",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Conditioning Zero Out"
+ }
+ },
+ "12": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "second.png"
+ },
+ "_meta": {
+ "title": "Second reference"
+ }
+ }
+}
diff --git a/apps/cloud-surf/app.json b/apps/cloud-surf/app.json
new file mode 100644
index 0000000..3b16fd2
--- /dev/null
+++ b/apps/cloud-surf/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "cloud-surf",
+ "name": "Cloud Surf",
+ "section": "video",
+ "summary": "You surfing across pink clouds at sunset.",
+ "higgsfield_feature": "Cloud Surf",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "The person surfs on a surfboard across fluffy pink clouds at sunset, carving through the cloud tops, spray of cloud mist, dynamic tracking shot; wind and a cheerful whoosh. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "16:9 (Widescreen)"
+ }
+}
diff --git a/apps/cloud-surf/workflow.api.json b/apps/cloud-surf/workflow.api.json
new file mode 100644
index 0000000..411a400
--- /dev/null
+++ b/apps/cloud-surf/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/cloud-surf",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/face-swap/app.json b/apps/face-swap/app.json
new file mode 100644
index 0000000..10a3696
--- /dev/null
+++ b/apps/face-swap/app.json
@@ -0,0 +1,134 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "face-swap",
+ "name": "Face Swap",
+ "section": "image",
+ "summary": "Puts the face from a second photo onto the first; for faces you own or have permission to use.",
+ "higgsfield_feature": "Face Swap",
+ "pack": "image-flux3",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_flux3_klein_distilled_image_edit.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/edit's (built by tools/build_apps.py) plus a second LoadImage on ReferenceMultiLatent's images.image1; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Photo to edit",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "help": "Only faces you own or have permission to use.",
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "image2",
+ "type": "image",
+ "label": "Photo of the face",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "12",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "4",
+ "field": "text",
+ "template": "The photo from <ref_image_0> with the face replaced by the face of the person in <ref_image_1>, matching the original head angle, expression, lighting and skin tone; everything else unchanged. {details}"
+ }
+ ]
+ },
+ {
+ "id": "steps",
+ "type": "slider",
+ "label": "Steps",
+ "min": 8,
+ "max": 32,
+ "step": 1,
+ "default": 16,
+ "bind": [
+ {
+ "node": "7",
+ "field": "steps"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "7",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "9",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "diffusion_models/flux3-image-klein-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 5.83,
+ "phase": "denoise",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "text_encoders/qwen3vl_4b_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 4.86,
+ "phase": "encode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "vae/flux2-vae.safetensors",
+ "dir": "vae",
+ "gb": 0.34,
+ "phase": "decode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ }
+ ],
+ "credits": [
+ "FLUX 3 Image Klein by Black Forest Labs (FLUX Kommunity License)"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "image2": "test/second.png"
+ }
+}
diff --git a/apps/face-swap/workflow.api.json b/apps/face-swap/workflow.api.json
new file mode 100644
index 0000000..be3697a
--- /dev/null
+++ b/apps/face-swap/workflow.api.json
@@ -0,0 +1,164 @@
+{
+ "1": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "flux3-image-klein-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "2": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen3vl_4b_int8_convrot.safetensors",
+ "type": "flux3",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "3": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "flux2-vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "4": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "2",
+ 0
+ ],
+ "text": "Change the jacket to deep red leather, keep everything else the same."
+ },
+ "_meta": {
+ "title": "Positive Prompt"
+ }
+ },
+ "7": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "1",
+ 0
+ ],
+ "positive": [
+ "17",
+ 0
+ ],
+ "negative": [
+ "17",
+ 1
+ ],
+ "latent_image": [
+ "17",
+ 2
+ ],
+ "seed": 563228241303057,
+ "steps": 16,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "7",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "9": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "8",
+ 0
+ ],
+ "filename_prefix": "freefield/face-swap"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "woman_gold_earring_teal_top_portrait.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "17": {
+ "class_type": "ReferenceMultiLatent",
+ "inputs": {
+ "positive": [
+ "4",
+ 0
+ ],
+ "negative": [
+ "32",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ],
+ "images.image0": [
+ "11",
+ 0
+ ],
+ "resolution": "resize",
+ "resolution.megapixels": 1,
+ "images.image1": [
+ "12",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Set Reference Multi Latent"
+ }
+ },
+ "32": {
+ "class_type": "ConditioningZeroOut",
+ "inputs": {
+ "conditioning": [
+ "4",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Conditioning Zero Out"
+ }
+ },
+ "12": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "second.png"
+ },
+ "_meta": {
+ "title": "Second reference"
+ }
+ }
+}
diff --git a/apps/fashion-factory/app.json b/apps/fashion-factory/app.json
new file mode 100644
index 0000000..bdb9c1a
--- /dev/null
+++ b/apps/fashion-factory/app.json
@@ -0,0 +1,157 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "fashion-factory",
+ "name": "Fashion Factory",
+ "section": "image",
+ "summary": "A fashion catalog photo of you wearing a garment from a second photo.",
+ "higgsfield_feature": "Fashion Factory",
+ "pack": "image-flux3",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_flux3_klein_distilled_image_edit.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/edit's (built by tools/build_apps.py) plus a second LoadImage on ReferenceMultiLatent's images.image1; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "image2",
+ "type": "image",
+ "label": "Garment photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "12",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "scene",
+ "type": "select",
+ "label": "Scene",
+ "default": "in a white studio cyclorama",
+ "options": [
+ {
+ "value": "in a white studio cyclorama",
+ "label": "Studio cyclorama"
+ },
+ {
+ "value": "on a city street at golden hour",
+ "label": "City street, golden hour"
+ },
+ {
+ "value": "in a minimalist loft",
+ "label": "Minimalist loft"
+ },
+ {
+ "value": "walking a runway under spotlights",
+ "label": "Runway"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "4",
+ "field": "text",
+ "template": "A fashion catalog photo: the person from <ref_image_0> wearing the garment from <ref_image_1>, full-length, {scene}. Keep the face identical and the garment's cut, color and pattern exactly as in its photo. {details}"
+ }
+ ]
+ },
+ {
+ "id": "steps",
+ "type": "slider",
+ "label": "Steps",
+ "min": 8,
+ "max": 32,
+ "step": 1,
+ "default": 16,
+ "bind": [
+ {
+ "node": "7",
+ "field": "steps"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "7",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "9",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "diffusion_models/flux3-image-klein-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 5.83,
+ "phase": "denoise",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "text_encoders/qwen3vl_4b_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 4.86,
+ "phase": "encode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "vae/flux2-vae.safetensors",
+ "dir": "vae",
+ "gb": 0.34,
+ "phase": "decode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ }
+ ],
+ "credits": [
+ "FLUX 3 Image Klein by Black Forest Labs (FLUX Kommunity License)"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "image2": "test/second.png"
+ }
+}
diff --git a/apps/fashion-factory/workflow.api.json b/apps/fashion-factory/workflow.api.json
new file mode 100644
index 0000000..8aa3888
--- /dev/null
+++ b/apps/fashion-factory/workflow.api.json
@@ -0,0 +1,164 @@
+{
+ "1": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "flux3-image-klein-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "2": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen3vl_4b_int8_convrot.safetensors",
+ "type": "flux3",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "3": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "flux2-vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "4": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "2",
+ 0
+ ],
+ "text": "Change the jacket to deep red leather, keep everything else the same."
+ },
+ "_meta": {
+ "title": "Positive Prompt"
+ }
+ },
+ "7": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "1",
+ 0
+ ],
+ "positive": [
+ "17",
+ 0
+ ],
+ "negative": [
+ "17",
+ 1
+ ],
+ "latent_image": [
+ "17",
+ 2
+ ],
+ "seed": 563228241303057,
+ "steps": 16,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "7",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "9": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "8",
+ 0
+ ],
+ "filename_prefix": "freefield/fashion-factory"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "woman_gold_earring_teal_top_portrait.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "17": {
+ "class_type": "ReferenceMultiLatent",
+ "inputs": {
+ "positive": [
+ "4",
+ 0
+ ],
+ "negative": [
+ "32",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ],
+ "images.image0": [
+ "11",
+ 0
+ ],
+ "resolution": "resize",
+ "resolution.megapixels": 1,
+ "images.image1": [
+ "12",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Set Reference Multi Latent"
+ }
+ },
+ "32": {
+ "class_type": "ConditioningZeroOut",
+ "inputs": {
+ "conditioning": [
+ "4",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Conditioning Zero Out"
+ }
+ },
+ "12": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "second.png"
+ },
+ "_meta": {
+ "title": "Second reference"
+ }
+ }
+}
diff --git a/apps/idol/app.json b/apps/idol/app.json
new file mode 100644
index 0000000..3fde22f
--- /dev/null
+++ b/apps/idol/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "idol",
+ "name": "Idol",
+ "section": "video",
+ "summary": "Your K-pop idol moment: spotlights, wind, confetti, a pose.",
+ "higgsfield_feature": "Idol",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "A pop idol stage moment: the person stands center stage under spotlights, hair blown by a fan, confetti falling, strikes a pose and smiles at the camera; crowd cheering and a pop beat. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "9:16 (Portrait Widescreen)"
+ }
+}
diff --git a/apps/idol/workflow.api.json b/apps/idol/workflow.api.json
new file mode 100644
index 0000000..d66a7c1
--- /dev/null
+++ b/apps/idol/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/idol",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/kick-ad/app.json b/apps/kick-ad/app.json
new file mode 100644
index 0000000..b237d2e
--- /dev/null
+++ b/apps/kick-ad/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "kick-ad",
+ "name": "Kick Ad",
+ "section": "video",
+ "summary": "Your product kicks the plain grey competition off screen.",
+ "higgsfield_feature": "Kick Ad",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Product photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "The product stands in the center; plain grey unmarked boxes slide in from the sides and the product knocks them away with a swift cartoon kick, they tumble off screen; comedic whoosh and thud sounds. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "16:9 (Widescreen)"
+ }
+}
diff --git a/apps/kick-ad/workflow.api.json b/apps/kick-ad/workflow.api.json
new file mode 100644
index 0000000..c0ea389
--- /dev/null
+++ b/apps/kick-ad/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/kick-ad",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/magic-button/app.json b/apps/magic-button/app.json
new file mode 100644
index 0000000..34accfb
--- /dev/null
+++ b/apps/magic-button/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "magic-button",
+ "name": "Magic Button",
+ "section": "video",
+ "summary": "A glowing button turns the scene around your product magical.",
+ "higgsfield_feature": "Magic button",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Product photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "A hand presses a big glowing button next to the product; the scene transforms into a magical world with floating sparks and soft light, the product levitating and turning slowly; a magical chime. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "9:16 (Portrait Widescreen)"
+ }
+}
diff --git a/apps/magic-button/workflow.api.json b/apps/magic-button/workflow.api.json
new file mode 100644
index 0000000..4c369c4
--- /dev/null
+++ b/apps/magic-button/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/magic-button",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/melting-doodle/app.json b/apps/melting-doodle/app.json
new file mode 100644
index 0000000..760962c
--- /dev/null
+++ b/apps/melting-doodle/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "melting-doodle",
+ "name": "Melting Doodle",
+ "section": "video",
+ "summary": "Cartoon doodles appear over your photo and melt down the frame.",
+ "higgsfield_feature": "Melting Doodle",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "Colorful cartoon doodles appear drawn over the scene, then begin to melt and drip down the frame like wet paint, dripping and merging; a soft squelching sound. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "9:16 (Portrait Widescreen)"
+ }
+}
diff --git a/apps/melting-doodle/workflow.api.json b/apps/melting-doodle/workflow.api.json
new file mode 100644
index 0000000..c498cc2
--- /dev/null
+++ b/apps/melting-doodle/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/melting-doodle",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/mukbang/app.json b/apps/mukbang/app.json
new file mode 100644
index 0000000..7d61402
--- /dev/null
+++ b/apps/mukbang/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "mukbang",
+ "name": "Mukbang",
+ "section": "video",
+ "summary": "You at a table full of food, eating for the camera.",
+ "higgsfield_feature": "Mukbang",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "A mukbang video: the person sits at a table loaded with food in front of a microphone, picks up a piece and eats with relish, nodding at the camera; crunching and chewing sounds, a short delighted hum. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "9:16 (Portrait Widescreen)"
+ }
+}
diff --git a/apps/mukbang/workflow.api.json b/apps/mukbang/workflow.api.json
new file mode 100644
index 0000000..3aad238
--- /dev/null
+++ b/apps/mukbang/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/mukbang",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/outfit-shot/app.json b/apps/outfit-shot/app.json
new file mode 100644
index 0000000..7133dba
--- /dev/null
+++ b/apps/outfit-shot/app.json
@@ -0,0 +1,153 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "outfit-shot",
+ "name": "Outfit Shot",
+ "section": "image",
+ "summary": "A studio fashion shot of a garment on a model from your photo.",
+ "higgsfield_feature": "Outfit Shot",
+ "pack": "image-flux3",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_flux3_klein_distilled_image_edit.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/edit's (built by tools/build_apps.py) plus a second LoadImage on ReferenceMultiLatent's images.image1; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Model photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "image2",
+ "type": "image",
+ "label": "Garment photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "12",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "pose",
+ "type": "select",
+ "label": "Pose",
+ "default": "standing straight, hands relaxed",
+ "options": [
+ {
+ "value": "standing straight, hands relaxed",
+ "label": "Standing"
+ },
+ {
+ "value": "mid-step, walking toward the camera",
+ "label": "Walking"
+ },
+ {
+ "value": "sitting on a wooden stool",
+ "label": "On a stool"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "4",
+ "field": "text",
+ "template": "A studio fashion photo of the person from <ref_image_0> modeling the garment from <ref_image_1>, {pose}, clean studio backdrop, soft light. Keep the face identical and the garment exactly as in its photo. {details}"
+ }
+ ]
+ },
+ {
+ "id": "steps",
+ "type": "slider",
+ "label": "Steps",
+ "min": 8,
+ "max": 32,
+ "step": 1,
+ "default": 16,
+ "bind": [
+ {
+ "node": "7",
+ "field": "steps"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "7",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "9",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "diffusion_models/flux3-image-klein-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 5.83,
+ "phase": "denoise",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "text_encoders/qwen3vl_4b_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 4.86,
+ "phase": "encode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "vae/flux2-vae.safetensors",
+ "dir": "vae",
+ "gb": 0.34,
+ "phase": "decode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ }
+ ],
+ "credits": [
+ "FLUX 3 Image Klein by Black Forest Labs (FLUX Kommunity License)"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "image2": "test/second.png"
+ }
+}
diff --git a/apps/outfit-shot/workflow.api.json b/apps/outfit-shot/workflow.api.json
new file mode 100644
index 0000000..f66f7fc
--- /dev/null
+++ b/apps/outfit-shot/workflow.api.json
@@ -0,0 +1,164 @@
+{
+ "1": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "flux3-image-klein-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "2": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen3vl_4b_int8_convrot.safetensors",
+ "type": "flux3",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "3": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "flux2-vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "4": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "2",
+ 0
+ ],
+ "text": "Change the jacket to deep red leather, keep everything else the same."
+ },
+ "_meta": {
+ "title": "Positive Prompt"
+ }
+ },
+ "7": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "1",
+ 0
+ ],
+ "positive": [
+ "17",
+ 0
+ ],
+ "negative": [
+ "17",
+ 1
+ ],
+ "latent_image": [
+ "17",
+ 2
+ ],
+ "seed": 563228241303057,
+ "steps": 16,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "7",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "9": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "8",
+ 0
+ ],
+ "filename_prefix": "freefield/outfit-shot"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "woman_gold_earring_teal_top_portrait.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "17": {
+ "class_type": "ReferenceMultiLatent",
+ "inputs": {
+ "positive": [
+ "4",
+ 0
+ ],
+ "negative": [
+ "32",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ],
+ "images.image0": [
+ "11",
+ 0
+ ],
+ "resolution": "resize",
+ "resolution.megapixels": 1,
+ "images.image1": [
+ "12",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Set Reference Multi Latent"
+ }
+ },
+ "32": {
+ "class_type": "ConditioningZeroOut",
+ "inputs": {
+ "conditioning": [
+ "4",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Conditioning Zero Out"
+ }
+ },
+ "12": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "second.png"
+ },
+ "_meta": {
+ "title": "Second reference"
+ }
+ }
+}
diff --git a/apps/outfit-swap/app.json b/apps/outfit-swap/app.json
new file mode 100644
index 0000000..c4bb5ea
--- /dev/null
+++ b/apps/outfit-swap/app.json
@@ -0,0 +1,133 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "outfit-swap",
+ "name": "Outfit Swap",
+ "section": "image",
+ "summary": "Wear the whole outfit of the person in another photo.",
+ "higgsfield_feature": "Outfit Swap",
+ "pack": "image-flux3",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_flux3_klein_distilled_image_edit.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/edit's (built by tools/build_apps.py) plus a second LoadImage on ReferenceMultiLatent's images.image1; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "image2",
+ "type": "image",
+ "label": "Photo of the outfit",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "12",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "4",
+ "field": "text",
+ "template": "The person from <ref_image_0> now dressed in the complete outfit worn by the person in <ref_image_1>, same pose, framing and background as <ref_image_0>. Keep the face identical. {details}"
+ }
+ ]
+ },
+ {
+ "id": "steps",
+ "type": "slider",
+ "label": "Steps",
+ "min": 8,
+ "max": 32,
+ "step": 1,
+ "default": 16,
+ "bind": [
+ {
+ "node": "7",
+ "field": "steps"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "7",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "9",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "diffusion_models/flux3-image-klein-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 5.83,
+ "phase": "denoise",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "text_encoders/qwen3vl_4b_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 4.86,
+ "phase": "encode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "vae/flux2-vae.safetensors",
+ "dir": "vae",
+ "gb": 0.34,
+ "phase": "decode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ }
+ ],
+ "credits": [
+ "FLUX 3 Image Klein by Black Forest Labs (FLUX Kommunity License)"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "image2": "test/second.png"
+ }
+}
diff --git a/apps/outfit-swap/workflow.api.json b/apps/outfit-swap/workflow.api.json
new file mode 100644
index 0000000..85feb62
--- /dev/null
+++ b/apps/outfit-swap/workflow.api.json
@@ -0,0 +1,164 @@
+{
+ "1": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "flux3-image-klein-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "2": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen3vl_4b_int8_convrot.safetensors",
+ "type": "flux3",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "3": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "flux2-vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "4": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "2",
+ 0
+ ],
+ "text": "Change the jacket to deep red leather, keep everything else the same."
+ },
+ "_meta": {
+ "title": "Positive Prompt"
+ }
+ },
+ "7": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "1",
+ 0
+ ],
+ "positive": [
+ "17",
+ 0
+ ],
+ "negative": [
+ "17",
+ 1
+ ],
+ "latent_image": [
+ "17",
+ 2
+ ],
+ "seed": 563228241303057,
+ "steps": 16,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "7",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "9": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "8",
+ 0
+ ],
+ "filename_prefix": "freefield/outfit-swap"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "woman_gold_earring_teal_top_portrait.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "17": {
+ "class_type": "ReferenceMultiLatent",
+ "inputs": {
+ "positive": [
+ "4",
+ 0
+ ],
+ "negative": [
+ "32",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ],
+ "images.image0": [
+ "11",
+ 0
+ ],
+ "resolution": "resize",
+ "resolution.megapixels": 1,
+ "images.image1": [
+ "12",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Set Reference Multi Latent"
+ }
+ },
+ "32": {
+ "class_type": "ConditioningZeroOut",
+ "inputs": {
+ "conditioning": [
+ "4",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Conditioning Zero Out"
+ }
+ },
+ "12": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "second.png"
+ },
+ "_meta": {
+ "title": "Second reference"
+ }
+ }
+}
diff --git a/apps/rap-video/app.json b/apps/rap-video/app.json
new file mode 100644
index 0000000..ec344c9
--- /dev/null
+++ b/apps/rap-video/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "rap-video",
+ "name": "Rap Video",
+ "section": "video",
+ "summary": "A rap music video of you: low angle, stage lights, smoke, a beat.",
+ "higgsfield_feature": "Rap God",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "A rap music video: the person performs toward the camera with confident hand gestures, low-angle shot, flashing stage lights and smoke, quick push-ins; a punchy hip-hop beat. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "9:16 (Portrait Widescreen)"
+ }
+}
diff --git a/apps/rap-video/workflow.api.json b/apps/rap-video/workflow.api.json
new file mode 100644
index 0000000..02bc156
--- /dev/null
+++ b/apps/rap-video/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/rap-video",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/roller-coaster/app.json b/apps/roller-coaster/app.json
new file mode 100644
index 0000000..b3cfe1d
--- /dev/null
+++ b/apps/roller-coaster/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "roller-coaster",
+ "name": "Roller Coaster",
+ "section": "video",
+ "summary": "You on a roller coaster, camera mounted on the car.",
+ "higgsfield_feature": "Roller Coaster",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "The person rides a roller coaster: the camera is mounted on the car facing them, the track climbs then drops, hair whipping, hands up, laughing with joy; clattering track and wind. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "9:16 (Portrait Widescreen)"
+ }
+}
diff --git a/apps/roller-coaster/workflow.api.json b/apps/roller-coaster/workflow.api.json
new file mode 100644
index 0000000..618ef25
--- /dev/null
+++ b/apps/roller-coaster/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/roller-coaster",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/shots/app.json b/apps/shots/app.json
new file mode 100644
index 0000000..141d8b4
--- /dev/null
+++ b/apps/shots/app.json
@@ -0,0 +1,162 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "shots",
+ "name": "Shots",
+ "section": "image",
+ "summary": "Nine framings of one image: wide, close-up, top-down, low angle and rotations.",
+ "higgsfield_feature": "Shots",
+ "pack": "image-qwen-edit",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_qwen_image_edit_2509_relight.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/relight's (built by tools/build_apps.py) with 镜头转换.safetensors in place of the Relight LoRA and no Relight trigger; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "shot",
+ "type": "select",
+ "label": "Shot",
+ "default": "将镜头转为广角镜头 Turn the camera to a wide-angle lens.",
+ "options": [
+ {
+ "value": "将镜头转为广角镜头 Turn the camera to a wide-angle lens.",
+ "label": "Wide-angle"
+ },
+ {
+ "value": "将镜头转为特写镜头 Turn the camera to a close-up.",
+ "label": "Close-up"
+ },
+ {
+ "value": "将镜头转为极特写镜头 Turn the camera to an extreme close-up.",
+ "label": "Extreme close-up"
+ },
+ {
+ "value": "将镜头转为俯视 Turn the camera to a top-down view.",
+ "label": "Top-down"
+ },
+ {
+ "value": "将镜头转为仰视 Turn the camera to a low-angle view looking up.",
+ "label": "Low angle"
+ },
+ {
+ "value": "将镜头向左旋转45度 Rotate the camera 45 degrees to the left.",
+ "label": "Rotate 45° left"
+ },
+ {
+ "value": "将镜头向右旋转45度 Rotate the camera 45 degrees to the right.",
+ "label": "Rotate 45° right"
+ },
+ {
+ "value": "将镜头向左旋转90度 Rotate the camera 90 degrees to the left.",
+ "label": "Rotate 90° left"
+ },
+ {
+ "value": "将镜头转到人物背后 Move the camera behind the subject.",
+ "label": "From behind"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "15",
+ "field": "value",
+ "template": "{shot} {details}"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "10:3",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "12",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Comfy-Org/Qwen-Image-Edit_ComfyUI",
+ "file": "split_files/diffusion_models/qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "dir": "diffusion_models",
+ "gb": 20.43,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "Comfy-Org/Qwen-Image_ComfyUI",
+ "file": "split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors",
+ "dir": "text_encoders",
+ "gb": 9.38,
+ "phase": "encode",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "Comfy-Org/Qwen-Image_ComfyUI",
+ "file": "split_files/vae/qwen_image_vae.safetensors",
+ "dir": "vae",
+ "gb": 0.25,
+ "phase": "decode",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "lightx2v/Qwen-Image-Lightning",
+ "file": "Qwen-Image-Edit-2509/Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors",
+ "dir": "loras",
+ "gb": 0.85,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "dx8152/Qwen-Edit-2509-Multiple-angles",
+ "file": "镜头转换.safetensors",
+ "dir": "loras",
+ "gb": 0.24,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ }
+ ],
+ "credits": [
+ "Qwen-Image-Edit-2509 by Alibaba Qwen (Apache-2.0)",
+ "Multiple Angles LoRA by dx8152 (Apache-2.0)",
+ "Lightning LoRA by lightx2v",
+ "Workflow: ComfyUI official template"
+ ],
+ "test": {
+ "image": "test/portrait.png"
+ }
+}
diff --git a/apps/shots/workflow.api.json b/apps/shots/workflow.api.json
new file mode 100644
index 0000000..f6a808b
--- /dev/null
+++ b/apps/shots/workflow.api.json
@@ -0,0 +1,251 @@
+{
+ "10:3": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "10:75",
+ 0
+ ],
+ "positive": [
+ "10:111",
+ 0
+ ],
+ "negative": [
+ "10:110",
+ 0
+ ],
+ "latent_image": [
+ "10:88",
+ 0
+ ],
+ "seed": 1118877715456453,
+ "steps": 4,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "10:8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "10:3",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "10:37": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "10:38": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen_2.5_vl_7b_fp8_scaled.safetensors",
+ "type": "qwen_image",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "10:39": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "qwen_image_vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "10:66": {
+ "class_type": "ModelSamplingAuraFlow",
+ "inputs": {
+ "model": [
+ "10:118",
+ 0
+ ],
+ "shift": 3
+ },
+ "_meta": {
+ "title": "ModelSamplingAuraFlow"
+ }
+ },
+ "10:75": {
+ "class_type": "CFGNorm",
+ "inputs": {
+ "model": [
+ "10:66",
+ 0
+ ],
+ "strength": 1,
+ "pre_cfg": false
+ },
+ "_meta": {
+ "title": "CFGNorm"
+ }
+ },
+ "10:88": {
+ "class_type": "VAEEncode",
+ "inputs": {
+ "pixels": [
+ "10:117",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Encode"
+ }
+ },
+ "10:89": {
+ "class_type": "LoraLoaderModelOnly",
+ "inputs": {
+ "model": [
+ "10:37",
+ 0
+ ],
+ "lora_name": "Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors",
+ "strength_model": 1
+ },
+ "_meta": {
+ "title": "Load LoRA"
+ }
+ },
+ "10:110": {
+ "class_type": "TextEncodeQwenImageEditPlus",
+ "inputs": {
+ "clip": [
+ "10:38",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ],
+ "image1": [
+ "10:117",
+ 0
+ ],
+ "prompt": ""
+ },
+ "_meta": {
+ "title": "Text Encode Qwen Image Edit Plus"
+ }
+ },
+ "10:111": {
+ "class_type": "TextEncodeQwenImageEditPlus",
+ "inputs": {
+ "clip": [
+ "10:38",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ],
+ "image1": [
+ "10:117",
+ 0
+ ],
+ "prompt": [
+ "14",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Text Encode Qwen Image Edit Plus"
+ }
+ },
+ "10:117": {
+ "class_type": "FluxKontextImageScale",
+ "inputs": {
+ "image": [
+ "11",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "FluxKontextImageScale"
+ }
+ },
+ "10:118": {
+ "class_type": "LoraLoaderModelOnly",
+ "inputs": {
+ "model": [
+ "10:89",
+ 0
+ ],
+ "lora_name": "镜头转换.safetensors",
+ "strength_model": 1
+ },
+ "_meta": {
+ "title": "Load LoRA"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "warm_living_room.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "12": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "10:8",
+ 0
+ ],
+ "filename_prefix": "freefield/shots"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "14": {
+ "class_type": "StringConcatenate",
+ "inputs": {
+ "string_b": [
+ "15",
+ 0
+ ],
+ "string_a": "",
+ "delimiter": ""
+ },
+ "_meta": {
+ "title": "LoRA trigger word"
+ }
+ },
+ "15": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": " relight the image using twilight light (with a warm color temperature, soft light, and blurred shadows)"
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ }
+}
diff --git a/apps/storm-creature/app.json b/apps/storm-creature/app.json
new file mode 100644
index 0000000..7b9a986
--- /dev/null
+++ b/apps/storm-creature/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "storm-creature",
+ "name": "Storm Creature",
+ "section": "video",
+ "summary": "A creature of clouds and lightning rises over your scene.",
+ "higgsfield_feature": "Storm Creature",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "A towering creature made of storm clouds and lightning rises from the clouds over the landscape, lightning flashing inside its body, rain starting; thunder and wind. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "16:9 (Widescreen)"
+ }
+}
diff --git a/apps/storm-creature/workflow.api.json b/apps/storm-creature/workflow.api.json
new file mode 100644
index 0000000..6652171
--- /dev/null
+++ b/apps/storm-creature/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/storm-creature",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/truck-ad/app.json b/apps/truck-ad/app.json
new file mode 100644
index 0000000..78e7fd2
--- /dev/null
+++ b/apps/truck-ad/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "truck-ad",
+ "name": "Truck Ad",
+ "section": "video",
+ "summary": "Your product printed huge on the side of a passing delivery truck.",
+ "higgsfield_feature": "Truck Ad",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Product photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "The camera pulls back from the product to reveal it printed huge on the side of a delivery truck driving along a city avenue, the truck passing by with sunlight flashing on it; traffic sound. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "16:9 (Widescreen)"
+ }
+}
diff --git a/apps/truck-ad/workflow.api.json b/apps/truck-ad/workflow.api.json
new file mode 100644
index 0000000..a1ad7ac
--- /dev/null
+++ b/apps/truck-ad/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/truck-ad",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/try-on/app.json b/apps/try-on/app.json
new file mode 100644
index 0000000..118b932
--- /dev/null
+++ b/apps/try-on/app.json
@@ -0,0 +1,133 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "try-on",
+ "name": "Virtual Try-On",
+ "section": "image",
+ "summary": "Try a garment on your own photo: same pose, same background, new clothes.",
+ "higgsfield_feature": "AI Stylist",
+ "pack": "image-flux3",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_flux3_klein_distilled_image_edit.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/edit's (built by tools/build_apps.py) plus a second LoadImage on ReferenceMultiLatent's images.image1; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "image2",
+ "type": "image",
+ "label": "Garment photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "12",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "4",
+ "field": "text",
+ "template": "The person from <ref_image_0> wearing the clothing item from <ref_image_1>, same pose, framing and background as <ref_image_0>, natural fit and folds. Keep the face and everything else identical; keep the garment's color and pattern. {details}"
+ }
+ ]
+ },
+ {
+ "id": "steps",
+ "type": "slider",
+ "label": "Steps",
+ "min": 8,
+ "max": 32,
+ "step": 1,
+ "default": 16,
+ "bind": [
+ {
+ "node": "7",
+ "field": "steps"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "7",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "9",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "diffusion_models/flux3-image-klein-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 5.83,
+ "phase": "denoise",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "text_encoders/qwen3vl_4b_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 4.86,
+ "phase": "encode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ },
+ {
+ "repo": "black-forest-labs/flux-3-image-klein-comfy",
+ "file": "vae/flux2-vae.safetensors",
+ "dir": "vae",
+ "gb": 0.34,
+ "phase": "decode",
+ "gated": true,
+ "license": "FLUX Kommunity License"
+ }
+ ],
+ "credits": [
+ "FLUX 3 Image Klein by Black Forest Labs (FLUX Kommunity License)"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "image2": "test/second.png"
+ }
+}
diff --git a/apps/try-on/workflow.api.json b/apps/try-on/workflow.api.json
new file mode 100644
index 0000000..aee2b8a
--- /dev/null
+++ b/apps/try-on/workflow.api.json
@@ -0,0 +1,164 @@
+{
+ "1": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "flux3-image-klein-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "2": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen3vl_4b_int8_convrot.safetensors",
+ "type": "flux3",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "3": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "flux2-vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "4": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "2",
+ 0
+ ],
+ "text": "Change the jacket to deep red leather, keep everything else the same."
+ },
+ "_meta": {
+ "title": "Positive Prompt"
+ }
+ },
+ "7": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "1",
+ 0
+ ],
+ "positive": [
+ "17",
+ 0
+ ],
+ "negative": [
+ "17",
+ 1
+ ],
+ "latent_image": [
+ "17",
+ 2
+ ],
+ "seed": 563228241303057,
+ "steps": 16,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "7",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "9": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "8",
+ 0
+ ],
+ "filename_prefix": "freefield/try-on"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "woman_gold_earring_teal_top_portrait.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "17": {
+ "class_type": "ReferenceMultiLatent",
+ "inputs": {
+ "positive": [
+ "4",
+ 0
+ ],
+ "negative": [
+ "32",
+ 0
+ ],
+ "vae": [
+ "3",
+ 0
+ ],
+ "images.image0": [
+ "11",
+ 0
+ ],
+ "resolution": "resize",
+ "resolution.megapixels": 1,
+ "images.image1": [
+ "12",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Set Reference Multi Latent"
+ }
+ },
+ "32": {
+ "class_type": "ConditioningZeroOut",
+ "inputs": {
+ "conditioning": [
+ "4",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Conditioning Zero Out"
+ }
+ },
+ "12": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "second.png"
+ },
+ "_meta": {
+ "title": "Second reference"
+ }
+ }
+}
diff --git a/apps/volcano-ad/app.json b/apps/volcano-ad/app.json
new file mode 100644
index 0000000..7db27d7
--- /dev/null
+++ b/apps/volcano-ad/app.json
@@ -0,0 +1,198 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "volcano-ad",
+ "name": "Volcano Ad",
+ "section": "video",
+ "summary": "Your product in front of an erupting volcano.",
+ "higgsfield_feature": "Volcano Ad",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "video_ltx2_5_i2v.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/video's (built by tools/build_apps.py); the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Product photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "395",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "398:376",
+ "field": "value",
+ "template": "The product stands on dark volcanic rock in front of an erupting volcano, lava fountains and glowing ash in the background, heat shimmer, slow camera push-in; a deep rumbling sound. {details}"
+ }
+ ]
+ },
+ {
+ "id": "duration",
+ "type": "slider",
+ "label": "Seconds",
+ "min": 2,
+ "max": 10,
+ "step": 1,
+ "default": 5,
+ "bind": [
+ {
+ "node": "398:362",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "enhance",
+ "type": "toggle",
+ "label": "Enhance prompt",
+ "default": true,
+ "help": "Rewrites your prompt with a small Gemma 4 model first",
+ "bind": [
+ {
+ "node": "398:383",
+ "field": "value"
+ }
+ ]
+ },
+ {
+ "id": "aspect",
+ "type": "select",
+ "label": "Aspect ratio",
+ "default": "16:9 (Widescreen)",
+ "options": [
+ {
+ "value": "16:9 (Widescreen)",
+ "label": "16:9"
+ },
+ {
+ "value": "9:16 (Portrait Widescreen)",
+ "label": "9:16"
+ },
+ {
+ "value": "1:1 (Square)",
+ "label": "1:1"
+ },
+ {
+ "value": "4:3 (Standard)",
+ "label": "4:3"
+ },
+ {
+ "value": "3:4 (Portrait Standard)",
+ "label": "3:4"
+ },
+ {
+ "value": "21:9 (Ultrawide)",
+ "label": "21:9"
+ }
+ ],
+ "bind": [
+ {
+ "node": "403",
+ "field": "aspect_ratio"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "398:339",
+ "field": "noise_seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "75",
+ "type": "video",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "diffusion_models/ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "dir": "diffusion_models",
+ "gb": 21.5,
+ "phase": "denoise",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "text_encoders/gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 15.37,
+ "phase": "encode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-video-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 1.47,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "vae/ltx-2.5-audio-vae-bf16.safetensors",
+ "dir": "vae",
+ "gb": 0.36,
+ "phase": "decode",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Lightricks/LTX-2.5",
+ "file": "latent_upscale_models/ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors",
+ "dir": "latent_upscale_models",
+ "gb": 1.0,
+ "phase": "aux",
+ "gated": true,
+ "license": "LTX-2.x Community License"
+ },
+ {
+ "repo": "Comfy-Org/gemma-4",
+ "file": "text_encoders/gemma4_e2b_it_int8_convrot.safetensors",
+ "dir": "text_encoders",
+ "gb": 5.2,
+ "phase": "encode",
+ "license": "Apache-2.0",
+ "optional": true
+ }
+ ],
+ "credits": [
+ "LTX-2.5 by Lightricks",
+ "Workflow adapted from ComfyUI's official LTX-2.5 template"
+ ],
+ "test": {
+ "image": "test/portrait.png",
+ "duration": 2,
+ "aspect": "16:9 (Widescreen)"
+ }
+}
diff --git a/apps/volcano-ad/workflow.api.json b/apps/volcano-ad/workflow.api.json
new file mode 100644
index 0000000..f5f9fab
--- /dev/null
+++ b/apps/volcano-ad/workflow.api.json
@@ -0,0 +1,728 @@
+{
+ "75": {
+ "class_type": "SaveVideo",
+ "inputs": {
+ "video": [
+ "398:370",
+ 0
+ ],
+ "filename_prefix": "freefield/volcano-ad",
+ "format": "auto",
+ "format.codec": "auto"
+ },
+ "_meta": {
+ "title": "Save Video"
+ }
+ },
+ "395": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "neon_cyborg_portrait.png"
+ },
+ "_meta": {
+ "title": "Load First Frame"
+ }
+ },
+ "398:338": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 42
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:339": {
+ "class_type": "RandomNoise",
+ "inputs": {
+ "noise_seed": 900359092894726
+ },
+ "_meta": {
+ "title": "RandomNoise"
+ }
+ },
+ "398:340": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:349",
+ 0
+ ],
+ "audio_latent": [
+ "398:367",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:341": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:344": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:339",
+ 0
+ ],
+ "guider": [
+ "398:388",
+ 0
+ ],
+ "sampler": [
+ "398:352",
+ 0
+ ],
+ "sigmas": [
+ "398:397",
+ 0
+ ],
+ "latent_image": [
+ "398:377",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:348": {
+ "class_type": "LTXVLatentUpsampler",
+ "inputs": {
+ "samples": [
+ "398:367",
+ 0
+ ],
+ "upscale_model": [
+ "398:371",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVLatentUpsampler"
+ }
+ },
+ "398:349": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:348",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 1
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:350": {
+ "class_type": "LTXVPreprocess",
+ "inputs": {
+ "image": [
+ "398:351",
+ 0
+ ],
+ "img_compression": 18
+ },
+ "_meta": {
+ "title": "LTXV Preprocess"
+ }
+ },
+ "398:351": {
+ "class_type": "ResizeImageMaskNode",
+ "inputs": {
+ "input": [
+ "395",
+ 0
+ ],
+ "resize_type": "scale longer dimension",
+ "resize_type.longer_size": 1536,
+ "scale_method": "lanczos"
+ },
+ "_meta": {
+ "title": "Resize Image/Mask"
+ }
+ },
+ "398:352": {
+ "class_type": "KSamplerSelect",
+ "inputs": {
+ "sampler_name": "euler_ancestral"
+ },
+ "_meta": {
+ "title": "KSamplerSelect"
+ }
+ },
+ "398:353": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:372",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:355": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:360",
+ 0
+ ],
+ "expression": "a/2"
+ },
+ "_meta": {
+ "title": "Math Expression"
+ }
+ },
+ "398:356": {
+ "class_type": "EmptyLTXVLatentVideo",
+ "inputs": {
+ "width": [
+ "398:353",
+ 1
+ ],
+ "height": [
+ "398:355",
+ 1
+ ],
+ "length": [
+ "398:378",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Video"
+ }
+ },
+ "398:357": {
+ "class_type": "LTXVImgToVideoInplace",
+ "inputs": {
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "latent": [
+ "398:356",
+ 0
+ ],
+ "bypass": [
+ "398:363",
+ 0
+ ],
+ "strength": 0.7
+ },
+ "_meta": {
+ "title": "LTXVImgToVideoInplace"
+ }
+ },
+ "398:358": {
+ "class_type": "LTXVAudioVAEDecode",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 1
+ ],
+ "audio_vae": [
+ "398:386",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXV Audio VAE Decode"
+ }
+ },
+ "398:359": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:361",
+ 0
+ ],
+ "expression": "a"
+ },
+ "_meta": {
+ "title": "Math Expression (fps)"
+ }
+ },
+ "398:360": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 1
+ ]
+ },
+ "_meta": {
+ "title": "Height"
+ }
+ },
+ "398:361": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 24
+ },
+ "_meta": {
+ "title": "Frame Rate"
+ }
+ },
+ "398:362": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": 5
+ },
+ "_meta": {
+ "title": "Duration"
+ }
+ },
+ "398:363": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": false
+ },
+ "_meta": {
+ "title": "Switch to Text to Video?"
+ }
+ },
+ "398:364": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": [
+ "398:382",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:365": {
+ "class_type": "LTXVConditioning",
+ "inputs": {
+ "positive": [
+ "398:364",
+ 0
+ ],
+ "negative": [
+ "398:373",
+ 0
+ ],
+ "frame_rate": [
+ "398:359",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "LTXVConditioning"
+ }
+ },
+ "398:366": {
+ "class_type": "LTXVEmptyLatentAudio",
+ "inputs": {
+ "audio_vae": [
+ "398:386",
+ 0
+ ],
+ "frames_number": [
+ "398:378",
+ 1
+ ],
+ "frame_rate": [
+ "398:359",
+ 1
+ ],
+ "batch_size": 1
+ },
+ "_meta": {
+ "title": "Empty LTXV Latent Audio"
+ }
+ },
+ "398:367": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:344",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:368": {
+ "class_type": "SamplerCustomAdvanced",
+ "inputs": {
+ "noise": [
+ "398:338",
+ 0
+ ],
+ "guider": [
+ "398:391",
+ 0
+ ],
+ "sampler": [
+ "398:341",
+ 0
+ ],
+ "sigmas": [
+ "398:396",
+ 0
+ ],
+ "latent_image": [
+ "398:340",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "SamplerCustomAdvanced"
+ }
+ },
+ "398:369": {
+ "class_type": "LTXVSeparateAVLatent",
+ "inputs": {
+ "av_latent": [
+ "398:368",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Separate AV Latent"
+ }
+ },
+ "398:370": {
+ "class_type": "CreateVideo",
+ "inputs": {
+ "images": [
+ "398:374",
+ 0
+ ],
+ "audio": [
+ "398:358",
+ 0
+ ],
+ "fps": [
+ "398:359",
+ 0
+ ],
+ "bit_depth": 8
+ },
+ "_meta": {
+ "title": "Create Video"
+ }
+ },
+ "398:371": {
+ "class_type": "LatentUpscaleModelLoader",
+ "inputs": {
+ "model_name": "ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors"
+ },
+ "_meta": {
+ "title": "Load Latent Upscale Model"
+ }
+ },
+ "398:372": {
+ "class_type": "PrimitiveInt",
+ "inputs": {
+ "value": [
+ "403",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Width"
+ }
+ },
+ "398:373": {
+ "class_type": "CLIPTextEncode",
+ "inputs": {
+ "clip": [
+ "398:387",
+ 0
+ ],
+ "text": "pc game, console game, video game, cartoon, childish, ugly"
+ },
+ "_meta": {
+ "title": "CLIP Text Encode (Prompt)"
+ }
+ },
+ "398:374": {
+ "class_type": "VAEDecodeTiled",
+ "inputs": {
+ "samples": [
+ "398:369",
+ 0
+ ],
+ "vae": [
+ "398:385",
+ 0
+ ],
+ "tile_size": 512,
+ "overlap": 64,
+ "temporal_size": 64,
+ "temporal_overlap": 16
+ },
+ "_meta": {
+ "title": "VAE Decode (Tiled)"
+ }
+ },
+ "398:376": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": "Egyptian royal in blue-and-gold headdress and high collar, white dress with golden embroidery and armbands, desert, robot soldiers in formation left and right. She walks steadily forward, head held level and gaze fixed ahead—no dipping or lowering of the head. The camera performs a single, smooth push-in only: starting in a wider shot of her, the robots, and the desert, it moves steadily forward until she is in a medium or medium-close frame, then holds. She stops, posture and head still upright, and says: “The old gods are silent. I am not.” Robot soldiers shift or march in place; sand and fabric move with the wind. No pull-back; the only camera move is the continuous push-in."
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ },
+ "398:377": {
+ "class_type": "LTXVConcatAVLatent",
+ "inputs": {
+ "video_latent": [
+ "398:357",
+ 0
+ ],
+ "audio_latent": [
+ "398:366",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Concat AV Latent"
+ }
+ },
+ "398:378": {
+ "class_type": "ComfyMathExpression",
+ "inputs": {
+ "values.a": [
+ "398:362",
+ 0
+ ],
+ "values.b": [
+ "398:361",
+ 0
+ ],
+ "expression": "a * b + 1"
+ },
+ "_meta": {
+ "title": "Math Expression (length)"
+ }
+ },
+ "398:380": {
+ "class_type": "TextGenerateLTX2Prompt",
+ "inputs": {
+ "clip": [
+ "398:393",
+ 0
+ ],
+ "image": [
+ "398:350",
+ 0
+ ],
+ "prompt": [
+ "398:376",
+ 0
+ ],
+ "max_length": 600,
+ "sampling_mode": "on",
+ "sampling_mode.temperature": 0.7,
+ "sampling_mode.top_k": 64,
+ "sampling_mode.top_p": 0.95,
+ "sampling_mode.min_p": 0.05,
+ "sampling_mode.repetition_penalty": 1.15,
+ "sampling_mode.seed": 0,
+ "sampling_mode.presence_penalty": 0,
+ "thinking": false,
+ "use_default_template": true
+ },
+ "_meta": {
+ "title": "Generate LTX2 Prompt"
+ }
+ },
+ "398:382": {
+ "class_type": "ComfySwitchNode",
+ "inputs": {
+ "on_false": [
+ "398:376",
+ 0
+ ],
+ "on_true": [
+ "398:380",
+ 0
+ ],
+ "switch": [
+ "398:383",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "If/Else Switch"
+ }
+ },
+ "398:383": {
+ "class_type": "PrimitiveBoolean",
+ "inputs": {
+ "value": true
+ },
+ "_meta": {
+ "title": "Boolean (Enable Prompt Enhance)"
+ }
+ },
+ "398:384": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "398:385": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-video-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:386": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "ltx-2.5-audio-vae-bf16.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "398:387": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:388": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:391": {
+ "class_type": "LTXVDualCFGGuider",
+ "inputs": {
+ "model": [
+ "398:384",
+ 0
+ ],
+ "positive": [
+ "398:365",
+ 0
+ ],
+ "negative": [
+ "398:365",
+ 1
+ ],
+ "video_cfg": 1,
+ "audio_cfg": 1
+ },
+ "_meta": {
+ "title": "LTXV Dual CFG Guider"
+ }
+ },
+ "398:393": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "gemma4_e2b_it_int8_convrot.safetensors",
+ "type": "ltxv",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "398:396": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "0.85, 0.7250, 0.4219, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "398:397": {
+ "class_type": "ManualSigmas",
+ "inputs": {
+ "sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
+ },
+ "_meta": {
+ "title": "ManualSigmas"
+ }
+ },
+ "403": {
+ "class_type": "ResolutionSelector",
+ "inputs": {
+ "aspect_ratio": "16:9 (Widescreen)",
+ "megapixels": 0.9,
+ "multiple": 32
+ },
+ "_meta": {
+ "title": "Resolution Selector"
+ }
+ }
+}
diff --git a/apps/whats-next/app.json b/apps/whats-next/app.json
new file mode 100644
index 0000000..f94b92d
--- /dev/null
+++ b/apps/whats-next/app.json
@@ -0,0 +1,158 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "whats-next",
+ "name": "What's Next",
+ "section": "image",
+ "summary": "The next shot of your scene: 8 cinematic continuations (next-scene LoRA on Qwen-Image-Edit).",
+ "higgsfield_feature": "What's Next?",
+ "pack": "image-qwen-edit",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_qwen_image_edit_2509_relight.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/relight's (built by tools/build_apps.py) with next-scene_lora-v2-3000.safetensors in place of the Relight LoRA and no Relight trigger; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "next",
+ "type": "select",
+ "label": "Continuation",
+ "default": "the camera pulls back to reveal the wider surroundings",
+ "options": [
+ {
+ "value": "the camera pulls back to reveal the wider surroundings",
+ "label": "Pull back to reveal"
+ },
+ {
+ "value": "the camera pushes in to a close-up on the main subject",
+ "label": "Push in to a close-up"
+ },
+ {
+ "value": "the camera pans right, revealing more of the scene",
+ "label": "Pan right"
+ },
+ {
+ "value": "a new character walks into the frame from the left",
+ "label": "Someone enters"
+ },
+ {
+ "value": "time passes and the light turns to dusk, lights coming on",
+ "label": "Dusk falls"
+ },
+ {
+ "value": "the weather turns and rain begins to fall",
+ "label": "Rain starts"
+ },
+ {
+ "value": "the camera tilts up toward the sky",
+ "label": "Tilt up to the sky"
+ },
+ {
+ "value": "the camera cuts to the reverse angle, looking back the other way",
+ "label": "Reverse angle"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "15",
+ "field": "value",
+ "template": "Next Scene: {next} {details}"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "10:3",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "12",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Comfy-Org/Qwen-Image-Edit_ComfyUI",
+ "file": "split_files/diffusion_models/qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "dir": "diffusion_models",
+ "gb": 20.43,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "Comfy-Org/Qwen-Image_ComfyUI",
+ "file": "split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors",
+ "dir": "text_encoders",
+ "gb": 9.38,
+ "phase": "encode",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "Comfy-Org/Qwen-Image_ComfyUI",
+ "file": "split_files/vae/qwen_image_vae.safetensors",
+ "dir": "vae",
+ "gb": 0.25,
+ "phase": "decode",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "lightx2v/Qwen-Image-Lightning",
+ "file": "Qwen-Image-Edit-2509/Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors",
+ "dir": "loras",
+ "gb": 0.85,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "lovis93/next-scene-qwen-image-lora-2509",
+ "file": "next-scene_lora-v2-3000.safetensors",
+ "dir": "loras",
+ "gb": 0.3,
+ "phase": "denoise",
+ "license": "MIT"
+ }
+ ],
+ "credits": [
+ "Qwen-Image-Edit-2509 by Alibaba Qwen (Apache-2.0)",
+ "next-scene LoRA v2 by lovis93 (MIT)",
+ "Lightning LoRA by lightx2v",
+ "Workflow: ComfyUI official template"
+ ],
+ "test": {
+ "image": "test/portrait.png"
+ }
+}
diff --git a/apps/whats-next/workflow.api.json b/apps/whats-next/workflow.api.json
new file mode 100644
index 0000000..1ed4077
--- /dev/null
+++ b/apps/whats-next/workflow.api.json
@@ -0,0 +1,251 @@
+{
+ "10:3": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "10:75",
+ 0
+ ],
+ "positive": [
+ "10:111",
+ 0
+ ],
+ "negative": [
+ "10:110",
+ 0
+ ],
+ "latent_image": [
+ "10:88",
+ 0
+ ],
+ "seed": 1118877715456453,
+ "steps": 4,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "10:8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "10:3",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "10:37": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "10:38": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen_2.5_vl_7b_fp8_scaled.safetensors",
+ "type": "qwen_image",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "10:39": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "qwen_image_vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "10:66": {
+ "class_type": "ModelSamplingAuraFlow",
+ "inputs": {
+ "model": [
+ "10:118",
+ 0
+ ],
+ "shift": 3
+ },
+ "_meta": {
+ "title": "ModelSamplingAuraFlow"
+ }
+ },
+ "10:75": {
+ "class_type": "CFGNorm",
+ "inputs": {
+ "model": [
+ "10:66",
+ 0
+ ],
+ "strength": 1,
+ "pre_cfg": false
+ },
+ "_meta": {
+ "title": "CFGNorm"
+ }
+ },
+ "10:88": {
+ "class_type": "VAEEncode",
+ "inputs": {
+ "pixels": [
+ "10:117",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Encode"
+ }
+ },
+ "10:89": {
+ "class_type": "LoraLoaderModelOnly",
+ "inputs": {
+ "model": [
+ "10:37",
+ 0
+ ],
+ "lora_name": "Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors",
+ "strength_model": 1
+ },
+ "_meta": {
+ "title": "Load LoRA"
+ }
+ },
+ "10:110": {
+ "class_type": "TextEncodeQwenImageEditPlus",
+ "inputs": {
+ "clip": [
+ "10:38",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ],
+ "image1": [
+ "10:117",
+ 0
+ ],
+ "prompt": ""
+ },
+ "_meta": {
+ "title": "Text Encode Qwen Image Edit Plus"
+ }
+ },
+ "10:111": {
+ "class_type": "TextEncodeQwenImageEditPlus",
+ "inputs": {
+ "clip": [
+ "10:38",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ],
+ "image1": [
+ "10:117",
+ 0
+ ],
+ "prompt": [
+ "14",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Text Encode Qwen Image Edit Plus"
+ }
+ },
+ "10:117": {
+ "class_type": "FluxKontextImageScale",
+ "inputs": {
+ "image": [
+ "11",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "FluxKontextImageScale"
+ }
+ },
+ "10:118": {
+ "class_type": "LoraLoaderModelOnly",
+ "inputs": {
+ "model": [
+ "10:89",
+ 0
+ ],
+ "lora_name": "next-scene_lora-v2-3000.safetensors",
+ "strength_model": 1
+ },
+ "_meta": {
+ "title": "Load LoRA"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "warm_living_room.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "12": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "10:8",
+ 0
+ ],
+ "filename_prefix": "freefield/whats-next"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "14": {
+ "class_type": "StringConcatenate",
+ "inputs": {
+ "string_b": [
+ "15",
+ 0
+ ],
+ "string_a": "",
+ "delimiter": ""
+ },
+ "_meta": {
+ "title": "LoRA trigger word"
+ }
+ },
+ "15": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": " relight the image using twilight light (with a warm color temperature, soft light, and blurred shadows)"
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ }
+}
diff --git a/apps/zooms/app.json b/apps/zooms/app.json
new file mode 100644
index 0000000..78f8c8e
--- /dev/null
+++ b/apps/zooms/app.json
@@ -0,0 +1,146 @@
+{
+ "$schema": "../../schema/app.schema.json",
+ "id": "zooms",
+ "name": "Zooms",
+ "section": "image",
+ "summary": "Zoom levels of one image, from a wide view to an extreme close-up.",
+ "higgsfield_feature": "Zooms",
+ "pack": "image-qwen-edit",
+ "status": "ready",
+ "workflow": "workflow.api.json",
+ "source": {
+ "template": "image_qwen_image_edit_2509_relight.json",
+ "comfyui": "0.42.0",
+ "notes": "Template app written by tools/template_apps.py: the workflow is apps/relight's (built by tools/build_apps.py) with 镜头转换.safetensors in place of the Relight LoRA and no Relight trigger; the prompt below is fixed and \"Extra direction\" is appended to it."
+ },
+ "inputs": [
+ {
+ "id": "image",
+ "type": "image",
+ "label": "Your photo",
+ "required": true,
+ "limits": {
+ "max_mb": 20
+ },
+ "bind": [
+ {
+ "node": "11",
+ "field": "image"
+ }
+ ]
+ },
+ {
+ "id": "zoom",
+ "type": "select",
+ "label": "Zoom",
+ "default": "将镜头转为特写镜头 Turn the camera to a close-up.",
+ "options": [
+ {
+ "value": "将镜头大幅向后移动并转为广角镜头 Move the camera far back and turn it to a wide-angle lens.",
+ "label": "Much wider"
+ },
+ {
+ "value": "将镜头向后移动 Move the camera backward.",
+ "label": "Wider"
+ },
+ {
+ "value": "将镜头向前移动 Move the camera forward.",
+ "label": "Closer"
+ },
+ {
+ "value": "将镜头转为特写镜头 Turn the camera to a close-up.",
+ "label": "Close-up"
+ },
+ {
+ "value": "将镜头转为极特写镜头 Turn the camera to an extreme close-up.",
+ "label": "Extreme close-up"
+ }
+ ]
+ },
+ {
+ "id": "details",
+ "type": "textarea",
+ "label": "Extra direction",
+ "default": "",
+ "help": "Added at the end of the app's prompt.",
+ "bind": [
+ {
+ "node": "15",
+ "field": "value",
+ "template": "{zoom} {details}"
+ }
+ ]
+ },
+ {
+ "id": "seed",
+ "type": "seed",
+ "label": "Seed",
+ "default": -1,
+ "help": "-1 picks a random seed",
+ "bind": [
+ {
+ "node": "10:3",
+ "field": "seed"
+ }
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "node": "12",
+ "type": "image",
+ "label": "Result"
+ }
+ ],
+ "models": [
+ {
+ "repo": "Comfy-Org/Qwen-Image-Edit_ComfyUI",
+ "file": "split_files/diffusion_models/qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "dir": "diffusion_models",
+ "gb": 20.43,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "Comfy-Org/Qwen-Image_ComfyUI",
+ "file": "split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors",
+ "dir": "text_encoders",
+ "gb": 9.38,
+ "phase": "encode",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "Comfy-Org/Qwen-Image_ComfyUI",
+ "file": "split_files/vae/qwen_image_vae.safetensors",
+ "dir": "vae",
+ "gb": 0.25,
+ "phase": "decode",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "lightx2v/Qwen-Image-Lightning",
+ "file": "Qwen-Image-Edit-2509/Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors",
+ "dir": "loras",
+ "gb": 0.85,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ },
+ {
+ "repo": "dx8152/Qwen-Edit-2509-Multiple-angles",
+ "file": "镜头转换.safetensors",
+ "dir": "loras",
+ "gb": 0.24,
+ "phase": "denoise",
+ "license": "Apache-2.0"
+ }
+ ],
+ "credits": [
+ "Qwen-Image-Edit-2509 by Alibaba Qwen (Apache-2.0)",
+ "Multiple Angles LoRA by dx8152 (Apache-2.0)",
+ "Lightning LoRA by lightx2v",
+ "Workflow: ComfyUI official template"
+ ],
+ "test": {
+ "image": "test/portrait.png"
+ }
+}
diff --git a/apps/zooms/workflow.api.json b/apps/zooms/workflow.api.json
new file mode 100644
index 0000000..3d1dade
--- /dev/null
+++ b/apps/zooms/workflow.api.json
@@ -0,0 +1,251 @@
+{
+ "10:3": {
+ "class_type": "KSampler",
+ "inputs": {
+ "model": [
+ "10:75",
+ 0
+ ],
+ "positive": [
+ "10:111",
+ 0
+ ],
+ "negative": [
+ "10:110",
+ 0
+ ],
+ "latent_image": [
+ "10:88",
+ 0
+ ],
+ "seed": 1118877715456453,
+ "steps": 4,
+ "cfg": 1,
+ "sampler_name": "euler",
+ "scheduler": "simple",
+ "denoise": 1
+ },
+ "_meta": {
+ "title": "KSampler"
+ }
+ },
+ "10:8": {
+ "class_type": "VAEDecode",
+ "inputs": {
+ "samples": [
+ "10:3",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Decode"
+ }
+ },
+ "10:37": {
+ "class_type": "UNETLoader",
+ "inputs": {
+ "unet_name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "weight_dtype": "default"
+ },
+ "_meta": {
+ "title": "Load Diffusion Model"
+ }
+ },
+ "10:38": {
+ "class_type": "CLIPLoader",
+ "inputs": {
+ "clip_name": "qwen_2.5_vl_7b_fp8_scaled.safetensors",
+ "type": "qwen_image",
+ "device": "default"
+ },
+ "_meta": {
+ "title": "Load CLIP"
+ }
+ },
+ "10:39": {
+ "class_type": "VAELoader",
+ "inputs": {
+ "vae_name": "qwen_image_vae.safetensors"
+ },
+ "_meta": {
+ "title": "Load VAE"
+ }
+ },
+ "10:66": {
+ "class_type": "ModelSamplingAuraFlow",
+ "inputs": {
+ "model": [
+ "10:118",
+ 0
+ ],
+ "shift": 3
+ },
+ "_meta": {
+ "title": "ModelSamplingAuraFlow"
+ }
+ },
+ "10:75": {
+ "class_type": "CFGNorm",
+ "inputs": {
+ "model": [
+ "10:66",
+ 0
+ ],
+ "strength": 1,
+ "pre_cfg": false
+ },
+ "_meta": {
+ "title": "CFGNorm"
+ }
+ },
+ "10:88": {
+ "class_type": "VAEEncode",
+ "inputs": {
+ "pixels": [
+ "10:117",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "VAE Encode"
+ }
+ },
+ "10:89": {
+ "class_type": "LoraLoaderModelOnly",
+ "inputs": {
+ "model": [
+ "10:37",
+ 0
+ ],
+ "lora_name": "Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors",
+ "strength_model": 1
+ },
+ "_meta": {
+ "title": "Load LoRA"
+ }
+ },
+ "10:110": {
+ "class_type": "TextEncodeQwenImageEditPlus",
+ "inputs": {
+ "clip": [
+ "10:38",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ],
+ "image1": [
+ "10:117",
+ 0
+ ],
+ "prompt": ""
+ },
+ "_meta": {
+ "title": "Text Encode Qwen Image Edit Plus"
+ }
+ },
+ "10:111": {
+ "class_type": "TextEncodeQwenImageEditPlus",
+ "inputs": {
+ "clip": [
+ "10:38",
+ 0
+ ],
+ "vae": [
+ "10:39",
+ 0
+ ],
+ "image1": [
+ "10:117",
+ 0
+ ],
+ "prompt": [
+ "14",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "Text Encode Qwen Image Edit Plus"
+ }
+ },
+ "10:117": {
+ "class_type": "FluxKontextImageScale",
+ "inputs": {
+ "image": [
+ "11",
+ 0
+ ]
+ },
+ "_meta": {
+ "title": "FluxKontextImageScale"
+ }
+ },
+ "10:118": {
+ "class_type": "LoraLoaderModelOnly",
+ "inputs": {
+ "model": [
+ "10:89",
+ 0
+ ],
+ "lora_name": "镜头转换.safetensors",
+ "strength_model": 1
+ },
+ "_meta": {
+ "title": "Load LoRA"
+ }
+ },
+ "11": {
+ "class_type": "LoadImage",
+ "inputs": {
+ "image": "warm_living_room.png"
+ },
+ "_meta": {
+ "title": "Load Image"
+ }
+ },
+ "12": {
+ "class_type": "SaveImage",
+ "inputs": {
+ "images": [
+ "10:8",
+ 0
+ ],
+ "filename_prefix": "freefield/zooms"
+ },
+ "_meta": {
+ "title": "Save Image"
+ }
+ },
+ "14": {
+ "class_type": "StringConcatenate",
+ "inputs": {
+ "string_b": [
+ "15",
+ 0
+ ],
+ "string_a": "",
+ "delimiter": ""
+ },
+ "_meta": {
+ "title": "LoRA trigger word"
+ }
+ },
+ "15": {
+ "class_type": "PrimitiveStringMultiline",
+ "inputs": {
+ "value": " relight the image using twilight light (with a warm color temperature, soft light, and blurred shadows)"
+ },
+ "_meta": {
+ "title": "Prompt"
+ }
+ }
+}
diff --git a/docs/vram-map.json b/docs/vram-map.json
index 1f15406..e672e46 100644
--- a/docs/vram-map.json
+++ b/docs/vram-map.json
@@ -252,6 +252,31 @@
"modal_gpu": "L4 / A10G (24 GB)",
"measured_peak_gb": null
},
+ {
+ "id": "character-swap",
+ "name": "Character Swap",
+ "section": "image",
+ "pack": "image-flux3",
+ "status": "ready",
+ "whole_gb": 11.03,
+ "phases_gb": {
+ "encode": 4.86,
+ "denoise": 5.83,
+ "denoise2": 0.0,
+ "decode": 0.34,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 5.83,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "flux3-image-klein-int8-convrot.safetensors",
+ "gb": 5.83
+ },
+ "verdict_24gb": "fits whole",
+ "modal_gpu": "L4 / A10G (24 GB)",
+ "measured_peak_gb": null
+ },
{
"id": "color-grade",
"name": "Color Grade",
@@ -377,6 +402,56 @@
"modal_gpu": "L4 / A10G (24 GB)",
"measured_peak_gb": null
},
+ {
+ "id": "face-swap",
+ "name": "Face Swap",
+ "section": "image",
+ "pack": "image-flux3",
+ "status": "ready",
+ "whole_gb": 11.03,
+ "phases_gb": {
+ "encode": 4.86,
+ "denoise": 5.83,
+ "denoise2": 0.0,
+ "decode": 0.34,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 5.83,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "flux3-image-klein-int8-convrot.safetensors",
+ "gb": 5.83
+ },
+ "verdict_24gb": "fits whole",
+ "modal_gpu": "L4 / A10G (24 GB)",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "fashion-factory",
+ "name": "Fashion Factory",
+ "section": "image",
+ "pack": "image-flux3",
+ "status": "ready",
+ "whole_gb": 11.03,
+ "phases_gb": {
+ "encode": 4.86,
+ "denoise": 5.83,
+ "denoise2": 0.0,
+ "decode": 0.34,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 5.83,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "flux3-image-klein-int8-convrot.safetensors",
+ "gb": 5.83
+ },
+ "verdict_24gb": "fits whole",
+ "modal_gpu": "L4 / A10G (24 GB)",
+ "measured_peak_gb": null
+ },
{
"id": "figure-3d",
"name": "3D Figure",
@@ -802,6 +877,56 @@
"modal_gpu": "L4 / A10G (24 GB)",
"measured_peak_gb": null
},
+ {
+ "id": "outfit-shot",
+ "name": "Outfit Shot",
+ "section": "image",
+ "pack": "image-flux3",
+ "status": "ready",
+ "whole_gb": 11.03,
+ "phases_gb": {
+ "encode": 4.86,
+ "denoise": 5.83,
+ "denoise2": 0.0,
+ "decode": 0.34,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 5.83,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "flux3-image-klein-int8-convrot.safetensors",
+ "gb": 5.83
+ },
+ "verdict_24gb": "fits whole",
+ "modal_gpu": "L4 / A10G (24 GB)",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "outfit-swap",
+ "name": "Outfit Swap",
+ "section": "image",
+ "pack": "image-flux3",
+ "status": "ready",
+ "whole_gb": 11.03,
+ "phases_gb": {
+ "encode": 4.86,
+ "denoise": 5.83,
+ "denoise2": 0.0,
+ "decode": 0.34,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 5.83,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "flux3-image-klein-int8-convrot.safetensors",
+ "gb": 5.83
+ },
+ "verdict_24gb": "fits whole",
+ "modal_gpu": "L4 / A10G (24 GB)",
+ "measured_peak_gb": null
+ },
{
"id": "outfit-vending",
"name": "Outfit Vending",
@@ -1102,6 +1227,31 @@
"modal_gpu": "L4 / A10G (24 GB)",
"measured_peak_gb": null
},
+ {
+ "id": "try-on",
+ "name": "Virtual Try-On",
+ "section": "image",
+ "pack": "image-flux3",
+ "status": "ready",
+ "whole_gb": 11.03,
+ "phases_gb": {
+ "encode": 4.86,
+ "denoise": 5.83,
+ "denoise2": 0.0,
+ "decode": 0.34,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 5.83,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "flux3-image-klein-int8-convrot.safetensors",
+ "gb": 5.83
+ },
+ "verdict_24gb": "fits whole",
+ "modal_gpu": "L4 / A10G (24 GB)",
+ "measured_peak_gb": null
+ },
{
"id": "ugc-shot",
"name": "UGC Shot",
@@ -1428,8 +1578,8 @@
"measured_peak_gb": null
},
{
- "id": "speak",
- "name": "Speak",
+ "id": "asmr-host",
+ "name": "ASMR Host",
"section": "video",
"pack": "video-ltx25",
"status": "ready",
@@ -1453,8 +1603,8 @@
"measured_peak_gb": null
},
{
- "id": "talking",
- "name": "Talking Avatar",
+ "id": "asmr-product",
+ "name": "ASMR Product",
"section": "video",
"pack": "video-ltx25",
"status": "ready",
@@ -1468,7 +1618,7 @@
},
"largest_phase": "denoise",
"largest_phase_gb": 21.5,
- "optional_gb": 5.63,
+ "optional_gb": 5.2,
"biggest_file": {
"name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
"gb": 21.5
@@ -1478,8 +1628,8 @@
"measured_peak_gb": null
},
{
- "id": "video",
- "name": "Image to Video",
+ "id": "asmr-whisper",
+ "name": "ASMR Whisper",
"section": "video",
"pack": "video-ltx25",
"status": "ready",
@@ -1503,8 +1653,8 @@
"measured_peak_gb": null
},
{
- "id": "video-t2v",
- "name": "Text to Video",
+ "id": "behind-the-scenes",
+ "name": "Behind the Scenes",
"section": "video",
"pack": "video-ltx25",
"status": "ready",
@@ -1528,21 +1678,521 @@
"measured_peak_gb": null
},
{
- "id": "relight",
- "name": "Relight",
- "section": "image",
- "pack": "image-qwen-edit",
+ "id": "chameleon",
+ "name": "Chameleon",
+ "section": "video",
+ "pack": "video-ltx25",
"status": "ready",
- "whole_gb": 31.15,
+ "whole_gb": 39.7,
"phases_gb": {
- "encode": 9.38,
- "denoise": 21.52,
+ "encode": 15.37,
+ "denoise": 21.5,
"denoise2": 0.0,
- "decode": 0.25,
- "aux": 0.0
+ "decode": 1.83,
+ "aux": 1.0
},
"largest_phase": "denoise",
- "largest_phase_gb": 21.52,
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "cloud-surf",
+ "name": "Cloud Surf",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "idol",
+ "name": "Idol",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "kick-ad",
+ "name": "Kick Ad",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "magic-button",
+ "name": "Magic Button",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "melting-doodle",
+ "name": "Melting Doodle",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "mukbang",
+ "name": "Mukbang",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "rap-video",
+ "name": "Rap Video",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "roller-coaster",
+ "name": "Roller Coaster",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "speak",
+ "name": "Speak",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "storm-creature",
+ "name": "Storm Creature",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "talking",
+ "name": "Talking Avatar",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.63,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "truck-ad",
+ "name": "Truck Ad",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "video",
+ "name": "Image to Video",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "video-t2v",
+ "name": "Text to Video",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "volcano-ad",
+ "name": "Volcano Ad",
+ "section": "video",
+ "pack": "video-ltx25",
+ "status": "ready",
+ "whole_gb": 39.7,
+ "phases_gb": {
+ "encode": 15.37,
+ "denoise": 21.5,
+ "denoise2": 0.0,
+ "decode": 1.83,
+ "aux": 1.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.5,
+ "optional_gb": 5.2,
+ "biggest_file": {
+ "name": "ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors",
+ "gb": 21.5
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "angles",
+ "name": "Angles",
+ "section": "image",
+ "pack": "image-qwen-edit",
+ "status": "ready",
+ "whole_gb": 31.15,
+ "phases_gb": {
+ "encode": 9.38,
+ "denoise": 21.52,
+ "denoise2": 0.0,
+ "decode": 0.25,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.52,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "gb": 20.43
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "relight",
+ "name": "Relight",
+ "section": "image",
+ "pack": "image-qwen-edit",
+ "status": "ready",
+ "whole_gb": 31.15,
+ "phases_gb": {
+ "encode": 9.38,
+ "denoise": 21.52,
+ "denoise2": 0.0,
+ "decode": 0.25,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.52,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "gb": 20.43
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "shots",
+ "name": "Shots",
+ "section": "image",
+ "pack": "image-qwen-edit",
+ "status": "ready",
+ "whole_gb": 31.15,
+ "phases_gb": {
+ "encode": 9.38,
+ "denoise": 21.52,
+ "denoise2": 0.0,
+ "decode": 0.25,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.52,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "gb": 20.43
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "zooms",
+ "name": "Zooms",
+ "section": "image",
+ "pack": "image-qwen-edit",
+ "status": "ready",
+ "whole_gb": 31.15,
+ "phases_gb": {
+ "encode": 9.38,
+ "denoise": 21.52,
+ "denoise2": 0.0,
+ "decode": 0.25,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.52,
+ "optional_gb": 0,
+ "biggest_file": {
+ "name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors",
+ "gb": 20.43
+ },
+ "verdict_24gb": "tight (by phases)",
+ "modal_gpu": "test on L4 / A10G; L40S (48 GB) is comfortable",
+ "measured_peak_gb": null
+ },
+ {
+ "id": "whats-next",
+ "name": "What's Next",
+ "section": "image",
+ "pack": "image-qwen-edit",
+ "status": "ready",
+ "whole_gb": 31.21,
+ "phases_gb": {
+ "encode": 9.38,
+ "denoise": 21.58,
+ "denoise2": 0.0,
+ "decode": 0.25,
+ "aux": 0.0
+ },
+ "largest_phase": "denoise",
+ "largest_phase_gb": 21.58,
"optional_gb": 0,
"biggest_file": {
"name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors",
diff --git a/docs/vram-map.md b/docs/vram-map.md
index abf9443..e20fa1b 100644
--- a/docs/vram-map.md
+++ b/docs/vram-map.md
@@ -18,11 +18,14 @@ When two apps run back to back with different models, the gateway asks ComfyUI t
| `burning-sunset` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `cafe-60s` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `character` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
+| `character-swap` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `color-grade` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `comic-book` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `commercial-faces` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `edit` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `edit-text` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
+| `face-swap` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
+| `fashion-factory` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `figure-3d` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `fridge-ad` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `game-styles` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
@@ -40,6 +43,8 @@ When two apps run back to back with different models, the gateway asks ComfyUI t
| `mascot` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `micro-beasts` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `mural` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
+| `outfit-shot` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
+| `outfit-swap` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `outfit-vending` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `packshot` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `paint-app` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
@@ -52,6 +57,7 @@ When two apps run back to back with different models, the gateway asks ComfyUI t
| `sketch-to-real` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `skin-enhancer` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `style-snap` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
+| `try-on` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `ugc-shot` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `victory-card` | ready | 11.0 GB | denoise 5.8 GB | flux3-image-klein-int8-convrot.safetensors (5.8) | fits whole | L4 / A10G (24 GB) | not measured |
| `cinema-studio` | ready | 12.2 GB | denoise 6.2 GB | z_image_turbo_int8_convrot.safetensors (6.2) | fits whole | L4 / A10G (24 GB) | not measured |
@@ -65,11 +71,31 @@ When two apps run back to back with different models, the gateway asks ComfyUI t
| `motion-transfer` | ready | 24.9 GB | denoise 16.6 GB | wan_animate_2_distill_int8_convrot.safetensors (16.6) | fits by phases | L4 / A10G (24 GB) | not measured |
| `effects` | ready | 25.0 GB (+13.0 optional) | denoise 16.8 GB | wan2.1_i2v_480p_14B_fp8_scaled.safetensors (16.4) | fits by phases | L4 / A10G (24 GB) | not measured |
| `recast` | ready | 28.6 GB | denoise 18.6 GB | wan2.1_14B_SCAIL_2_int8_convrot.safetensors (16.6) | fits by phases | L4 / A10G (24 GB) | not measured |
+| `asmr-host` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `asmr-product` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `asmr-whisper` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `behind-the-scenes` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `chameleon` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `cloud-surf` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `idol` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `kick-ad` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `magic-button` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `melting-doodle` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `mukbang` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `rap-video` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `roller-coaster` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
| `speak` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `storm-creature` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
| `talking` | ready | 39.7 GB (+5.6 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `truck-ad` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
| `video` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
| `video-t2v` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `volcano-ad` | ready | 39.7 GB (+5.2 optional) | denoise 21.5 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `angles` | ready | 31.1 GB | denoise 21.5 GB | qwen_image_edit_2509_fp8_e4m3fn.safetensors (20.4) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
| `relight` | ready | 31.1 GB | denoise 21.5 GB | qwen_image_edit_2509_fp8_e4m3fn.safetensors (20.4) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `shots` | ready | 31.1 GB | denoise 21.5 GB | qwen_image_edit_2509_fp8_e4m3fn.safetensors (20.4) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `zooms` | ready | 31.1 GB | denoise 21.5 GB | qwen_image_edit_2509_fp8_e4m3fn.safetensors (20.4) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
+| `whats-next` | ready | 31.2 GB | denoise 21.6 GB | qwen_image_edit_2509_fp8_e4m3fn.safetensors (20.4) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
| `vfx-clean-plate` | ready | 39.0 GB (+5.2 optional) | denoise 21.8 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
| `vfx-day-to-night` | ready | 39.0 GB (+5.2 optional) | denoise 21.8 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
| `transitions` | ready | 39.1 GB (+5.2 optional) | denoise 21.9 GB | ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors (21.5) | tight (by phases) | test on L4 / A10G; L40S (48 GB) is comfortable | not measured |
diff --git a/lab/cases.json b/lab/cases.json
index e2a1623..b094320 100644
--- a/lab/cases.json
+++ b/lab/cases.json
@@ -229,6 +229,8 @@
"note": "Reference sketch for Sketch to Real."},
{"id": "ref-tpl-sign", "app": "z-image", "values": {"prompt": "Photo of a small bakery storefront with a wooden sign above the door that reads OPEN in bold white letters, daylight, photorealistic", "width": 1280, "height": 768},
"note": "Reference with legible text for Edit Text."},
+ {"id": "ref-tpl-garment", "app": "z-image", "values": {"prompt": "Product photo of a mustard yellow knit cardigan laid flat on a plain white background, top-down, soft even light", "width": 1024, "height": 1024},
+ "note": "A garment on its own, for the two-reference apps."},
{"id": "tpl-poster", "showcase": true, "app": "poster", "values": {"image": "@case:ref-product", "headline": "NOIR"},
"note": "Poster. A poster layout with the bottle intact and the headline legible."},
@@ -309,6 +311,58 @@
{"id": "tpl-image-effects", "showcase": true, "app": "image-effects", "values": {"image": "@case:ref-fx-cafe", "effect": "glitch distortion with RGB split"},
"note": "Image Effects. RGB split and glitch bands over the cafe."},
{"id": "tpl-color-grade", "showcase": true, "app": "color-grade", "values": {"image": "@case:ref-fx-cafe", "look": "teal and orange blockbuster"},
- "note": "Color Grading. Only the colors change."}
+ "note": "Color Grading. Only the colors change."},
+ {"id": "tpl-fashion-factory", "showcase": true, "app": "fashion-factory", "values": {"image": "@case:ref-fx-woman-a", "image2": "@case:ref-tpl-garment"},
+ "note": "Fashion Factory. She wears the yellow cardigan; catalog framing."},
+ {"id": "tpl-try-on", "showcase": true, "app": "try-on", "values": {"image": "@case:ref-fx-man-b", "image2": "@case:ref-tpl-garment"},
+ "note": "AI Stylist. Same photo of him, now in the cardigan."},
+ {"id": "tpl-outfit-swap", "showcase": true, "app": "outfit-swap", "values": {"image": "@case:ref-face-a", "image2": "@case:ref-character"},
+ "note": "Outfit Swap. She wears his red jacket, jeans and sneakers."},
+ {"id": "tpl-outfit-shot", "showcase": true, "app": "outfit-shot", "values": {"image": "@case:ref-fx-woman-b", "image2": "@case:ref-tpl-garment"},
+ "note": "Outfit Shot. Studio shot in the cardigan, same face."},
+ {"id": "tpl-character-swap", "showcase": true, "app": "character-swap", "values": {"image": "@case:ref-fx-cafe", "image2": "@case:ref-fx-woman-a"},
+ "note": "Character Swap. The woman sits in the cafe in the man's place and pose."},
+ {"id": "tpl-face-swap", "showcase": true, "app": "face-swap", "values": {"image": "@case:ref-face-a", "image2": "@case:ref-fx-woman-b"},
+ "note": "Face Swap. Her portrait with the other woman's face; lighting matched."},
+ {"id": "tpl-behind-the-scenes", "showcase": true, "app": "behind-the-scenes", "values": {"image": "@case:ref-face-a", "aspect": "16:9 (Widescreen)"},
+ "note": "Behind the Scenes. A set appears around her as the camera widens."},
+ {"id": "tpl-asmr-product", "showcase": true, "app": "asmr-product", "values": {"image": "@case:ref-product", "aspect": "9:16 (Portrait Widescreen)"},
+ "note": "ASMR Add-On. Hands tap the bottle; tapping sounds."},
+ {"id": "tpl-asmr-whisper", "showcase": true, "app": "asmr-whisper", "values": {"image": "@case:ref-fx-woman-a", "aspect": "9:16 (Portrait Widescreen)"},
+ "note": "ASMR Classic. Whispering and tapping; calm."},
+ {"id": "tpl-asmr-host", "showcase": true, "app": "asmr-host", "values": {"image": "@case:ref-fx-man-b", "aspect": "9:16 (Portrait Widescreen)"},
+ "note": "ASMR Host. Studio, mic, whisper."},
+ {"id": "tpl-chameleon", "showcase": true, "app": "chameleon", "values": {"image": "@case:ref-fx-sneaker", "aspect": "16:9 (Widescreen)"},
+ "note": "Chameleon. The chameleon turns red and white like the sneaker."},
+ {"id": "tpl-magic-button", "showcase": true, "app": "magic-button", "values": {"image": "@case:ref-product", "aspect": "9:16 (Portrait Widescreen)"},
+ "note": "Magic button. Button, sparks, the bottle floats."},
+ {"id": "tpl-truck-ad", "showcase": true, "app": "truck-ad", "values": {"image": "@case:ref-product", "aspect": "16:9 (Widescreen)"},
+ "note": "Truck Ad. The bottle on a truck's side, truck moving."},
+ {"id": "tpl-kick-ad", "showcase": true, "app": "kick-ad", "values": {"image": "@case:ref-product", "aspect": "16:9 (Widescreen)"},
+ "note": "Kick Ad. Grey boxes tumble; cartoon sounds. No real brands."},
+ {"id": "tpl-volcano-ad", "showcase": true, "app": "volcano-ad", "values": {"image": "@case:ref-product", "aspect": "16:9 (Widescreen)"},
+ "note": "Volcano Ad. Lava behind the bottle; rumble."},
+ {"id": "tpl-roller-coaster", "showcase": true, "app": "roller-coaster", "values": {"image": "@case:ref-fx-woman-a", "aspect": "9:16 (Portrait Widescreen)"},
+ "note": "Roller Coaster. Climb, drop, hair flying."},
+ {"id": "tpl-storm-creature", "showcase": true, "app": "storm-creature", "values": {"image": "@case:ref-fx-storm", "aspect": "16:9 (Widescreen)"},
+ "note": "Storm Creature. A cloud giant with lightning inside."},
+ {"id": "tpl-cloud-surf", "showcase": true, "app": "cloud-surf", "values": {"image": "@case:ref-fx-man-b", "aspect": "16:9 (Widescreen)"},
+ "note": "Cloud Surf. Surfboard on clouds; same face."},
+ {"id": "tpl-mukbang", "showcase": true, "app": "mukbang", "values": {"image": "@case:ref-fx-man-b", "aspect": "9:16 (Portrait Widescreen)"},
+ "note": "Mukbang. Food, mic, eating sounds."},
+ {"id": "tpl-melting-doodle", "showcase": true, "app": "melting-doodle", "values": {"image": "@case:ref-fx-street", "aspect": "9:16 (Portrait Widescreen)"},
+ "note": "Melting Doodle. Doodles drawn, then dripping."},
+ {"id": "tpl-rap-video", "showcase": true, "app": "rap-video", "values": {"image": "@case:ref-fx-man-b", "aspect": "9:16 (Portrait Widescreen)"},
+ "note": "Rap God. Performance energy; a beat plays."},
+ {"id": "tpl-idol", "showcase": true, "app": "idol", "values": {"image": "@case:ref-fx-woman-a", "aspect": "9:16 (Portrait Widescreen)"},
+ "note": "Idol. Stage, confetti, pose."},
+ {"id": "tpl-angles", "showcase": true, "app": "angles", "values": {"image": "@case:ref-fx-cafe"},
+ "note": "Angles 2.0. The cafe seen from 45° to the left; same people and table."},
+ {"id": "tpl-shots", "showcase": true, "app": "shots", "values": {"image": "@case:ref-fx-lake"},
+ "note": "Shots. The lake scene as a wide-angle shot."},
+ {"id": "tpl-zooms", "showcase": true, "app": "zooms", "values": {"image": "@case:ref-fx-street"},
+ "note": "Zooms. A close-up of the woman in the raincoat."},
+ {"id": "tpl-whats-next", "showcase": true, "app": "whats-next", "values": {"image": "@case:ref-fx-road"},
+ "note": "What's Next?. The next shot of the road: wider, more landscape."}
]
}
diff --git a/packs/image-flux3.json b/packs/image-flux3.json
index c71fe0b..b910d3c 100644
--- a/packs/image-flux3.json
+++ b/packs/image-flux3.json
@@ -7,11 +7,14 @@
"burning-sunset",
"cafe-60s",
"character",
+ "character-swap",
"color-grade",
"comic-book",
"commercial-faces",
"edit",
"edit-text",
+ "face-swap",
+ "fashion-factory",
"figure-3d",
"fridge-ad",
"game-styles",
@@ -29,6 +32,8 @@
"mascot",
"micro-beasts",
"mural",
+ "outfit-shot",
+ "outfit-swap",
"outfit-vending",
"packshot",
"paint-app",
@@ -41,6 +46,7 @@
"sketch-to-real",
"skin-enhancer",
"style-snap",
+ "try-on",
"ugc-shot",
"victory-card"
],
@@ -63,11 +69,14 @@
"burning-sunset",
"cafe-60s",
"character",
+ "character-swap",
"color-grade",
"comic-book",
"commercial-faces",
"edit",
"edit-text",
+ "face-swap",
+ "fashion-factory",
"figure-3d",
"fridge-ad",
"game-styles",
@@ -85,6 +94,8 @@
"mascot",
"micro-beasts",
"mural",
+ "outfit-shot",
+ "outfit-swap",
"outfit-vending",
"packshot",
"paint-app",
@@ -97,6 +108,7 @@
"sketch-to-real",
"skin-enhancer",
"style-snap",
+ "try-on",
"ugc-shot",
"victory-card"
]
@@ -116,11 +128,14 @@
"burning-sunset",
"cafe-60s",
"character",
+ "character-swap",
"color-grade",
"comic-book",
"commercial-faces",
"edit",
"edit-text",
+ "face-swap",
+ "fashion-factory",
"figure-3d",
"fridge-ad",
"game-styles",
@@ -138,6 +153,8 @@
"mascot",
"micro-beasts",
"mural",
+ "outfit-shot",
+ "outfit-swap",
"outfit-vending",
"packshot",
"paint-app",
@@ -150,6 +167,7 @@
"sketch-to-real",
"skin-enhancer",
"style-snap",
+ "try-on",
"ugc-shot",
"victory-card"
]
@@ -169,11 +187,14 @@
"burning-sunset",
"cafe-60s",
"character",
+ "character-swap",
"color-grade",
"comic-book",
"commercial-faces",
"edit",
"edit-text",
+ "face-swap",
+ "fashion-factory",
"figure-3d",
"fridge-ad",
"game-styles",
@@ -191,6 +212,8 @@
"mascot",
"micro-beasts",
"mural",
+ "outfit-shot",
+ "outfit-swap",
"outfit-vending",
"packshot",
"paint-app",
@@ -203,6 +226,7 @@
"sketch-to-real",
"skin-enhancer",
"style-snap",
+ "try-on",
"ugc-shot",
"victory-card"
]
diff --git a/packs/image-qwen-edit.json b/packs/image-qwen-edit.json
index a6ebcc2..68b37c9 100644
--- a/packs/image-qwen-edit.json
+++ b/packs/image-qwen-edit.json
@@ -1,9 +1,13 @@
{
"pack": "image-qwen-edit",
"apps": [
- "relight"
+ "angles",
+ "relight",
+ "shots",
+ "whats-next",
+ "zooms"
],
- "download_gb": 31.15,
+ "download_gb": 31.69,
"needs_hf_token": false,
"custom_nodes": [],
"models": [
@@ -16,7 +20,11 @@
"gated": false,
"license": "Apache-2.0",
"used_by": [
- "relight"
+ "angles",
+ "relight",
+ "shots",
+ "whats-next",
+ "zooms"
]
},
{
@@ -28,7 +36,11 @@
"gated": false,
"license": "Apache-2.0",
"used_by": [
- "relight"
+ "angles",
+ "relight",
+ "shots",
+ "whats-next",
+ "zooms"
]
},
{
@@ -43,6 +55,32 @@
"relight"
]
},
+ {
+ "repo": "lovis93/next-scene-qwen-image-lora-2509",
+ "file": "next-scene_lora-v2-3000.safetensors",
+ "dir": "loras",
+ "gb": 0.3,
+ "name": "next-scene_lora-v2-3000.safetensors",
+ "gated": false,
+ "license": "MIT",
+ "used_by": [
+ "whats-next"
+ ]
+ },
+ {
+ "repo": "dx8152/Qwen-Edit-2509-Multiple-angles",
+ "file": "镜头转换.safetensors",
+ "dir": "loras",
+ "gb": 0.24,
+ "name": "镜头转换.safetensors",
+ "gated": false,
+ "license": "Apache-2.0",
+ "used_by": [
+ "angles",
+ "shots",
+ "zooms"
+ ]
+ },
{
"repo": "Comfy-Org/Qwen-Image_ComfyUI",
"file": "split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors",
@@ -52,7 +90,11 @@
"gated": false,
"license": "Apache-2.0",
"used_by": [
- "relight"
+ "angles",
+ "relight",
+ "shots",
+ "whats-next",
+ "zooms"
]
},
{
@@ -64,7 +106,11 @@
"gated": false,
"license": "Apache-2.0",
"used_by": [
- "relight"
+ "angles",
+ "relight",
+ "shots",
+ "whats-next",
+ "zooms"
]
}
]
diff --git a/packs/index.json b/packs/index.json
index 50255e7..9995efa 100644
--- a/packs/index.json
+++ b/packs/index.json
@@ -32,11 +32,14 @@
"burning-sunset",
"cafe-60s",
"character",
+ "character-swap",
"color-grade",
"comic-book",
"commercial-faces",
"edit",
"edit-text",
+ "face-swap",
+ "fashion-factory",
"figure-3d",
"fridge-ad",
"game-styles",
@@ -54,6 +57,8 @@
"mascot",
"micro-beasts",
"mural",
+ "outfit-shot",
+ "outfit-swap",
"outfit-vending",
"packshot",
"paint-app",
@@ -66,6 +71,7 @@
"sketch-to-real",
"skin-enhancer",
"style-snap",
+ "try-on",
"ugc-shot",
"victory-card"
],
@@ -82,9 +88,13 @@
},
"image-qwen-edit": {
"apps": [
- "relight"
+ "angles",
+ "relight",
+ "shots",
+ "whats-next",
+ "zooms"
],
- "download_gb": 31.15,
+ "download_gb": 31.69,
"needs_hf_token": false
},
"upscale": {
@@ -106,15 +116,31 @@
},
"video-ltx25": {
"apps": [
+ "asmr-host",
+ "asmr-product",
+ "asmr-whisper",
+ "behind-the-scenes",
"camera-copy",
+ "chameleon",
+ "cloud-surf",
+ "idol",
+ "kick-ad",
+ "magic-button",
+ "melting-doodle",
+ "mukbang",
+ "rap-video",
+ "roller-coaster",
"speak",
+ "storm-creature",
"talking",
"transitions",
+ "truck-ad",
"vfx-clean-plate",
"vfx-day-to-night",
"vfx-water",
"video",
- "video-t2v"
+ "video-t2v",
+ "volcano-ad"
],
"download_gb": 47.94,
"needs_hf_token": true
diff --git a/packs/video-ltx25.json b/packs/video-ltx25.json
index 8673c7e..da5b80c 100644
--- a/packs/video-ltx25.json
+++ b/packs/video-ltx25.json
@@ -1,15 +1,31 @@
{
"pack": "video-ltx25",
"apps": [
+ "asmr-host",
+ "asmr-product",
+ "asmr-whisper",
+ "behind-the-scenes",
"camera-copy",
+ "chameleon",
+ "cloud-surf",
+ "idol",
+ "kick-ad",
+ "magic-button",
+ "melting-doodle",
+ "mukbang",
+ "rap-video",
+ "roller-coaster",
"speak",
+ "storm-creature",
"talking",
"transitions",
+ "truck-ad",
"vfx-clean-plate",
"vfx-day-to-night",
"vfx-water",
"video",
- "video-t2v"
+ "video-t2v",
+ "volcano-ad"
],
"download_gb": 47.94,
"needs_hf_token": true,
@@ -29,15 +45,31 @@
"gated": true,
"license": "LTX-2.x Community License",
"used_by": [
+ "asmr-host",
+ "asmr-product",
+ "asmr-whisper",
+ "behind-the-scenes",
"camera-copy",
+ "chameleon",
+ "cloud-surf",
+ "idol",
+ "kick-ad",
+ "magic-button",
+ "melting-doodle",
+ "mukbang",
+ "rap-video",
+ "roller-coaster",
"speak",
+ "storm-creature",
"talking",
"transitions",
+ "truck-ad",
"vfx-clean-plate",
"vfx-day-to-night",
"vfx-water",
"video",
- "video-t2v"
+ "video-t2v",
+ "volcano-ad"
]
},
{
@@ -49,10 +81,26 @@
"gated": true,
"license": "LTX-2.x Community License",
"used_by": [
+ "asmr-host",
+ "asmr-product",
+ "asmr-whisper",
+ "behind-the-scenes",
+ "chameleon",
+ "cloud-surf",
+ "idol",
+ "kick-ad",
+ "magic-button",
+ "melting-doodle",
+ "mukbang",
+ "rap-video",
+ "roller-coaster",
"speak",
+ "storm-creature",
"talking",
+ "truck-ad",
"video",
- "video-t2v"
+ "video-t2v",
+ "volcano-ad"
]
},
{
@@ -136,15 +184,31 @@
"gated": true,
"license": "LTX-2.x Community License",
"used_by": [
+ "asmr-host",
+ "asmr-product",
+ "asmr-whisper",
+ "behind-the-scenes",
"camera-copy",
+ "chameleon",
+ "cloud-surf",
+ "idol",
+ "kick-ad",
+ "magic-button",
+ "melting-doodle",
+ "mukbang",
+ "rap-video",
+ "roller-coaster",
"speak",
+ "storm-creature",
"talking",
"transitions",
+ "truck-ad",
"vfx-clean-plate",
"vfx-day-to-night",
"vfx-water",
"video",
- "video-t2v"
+ "video-t2v",
+ "volcano-ad"
]
},
{
@@ -156,15 +220,31 @@
"gated": false,
"license": "Apache-2.0",
"used_by": [
+ "asmr-host",
+ "asmr-product",
+ "asmr-whisper",
+ "behind-the-scenes",
"camera-copy",
+ "chameleon",
+ "cloud-surf",
+ "idol",
+ "kick-ad",
+ "magic-button",
+ "melting-doodle",
+ "mukbang",
+ "rap-video",
+ "roller-coaster",
"speak",
+ "storm-creature",
"talking",
"transitions",
+ "truck-ad",
"vfx-clean-plate",
"vfx-day-to-night",
"vfx-water",
"video",
- "video-t2v"
+ "video-t2v",
+ "volcano-ad"
]
},
{
@@ -176,15 +256,31 @@
"gated": true,
"license": "LTX-2.x Community License",
"used_by": [
+ "asmr-host",
+ "asmr-product",
+ "asmr-whisper",
+ "behind-the-scenes",
"camera-copy",
+ "chameleon",
+ "cloud-surf",
+ "idol",
+ "kick-ad",
+ "magic-button",
+ "melting-doodle",
+ "mukbang",
+ "rap-video",
+ "roller-coaster",
"speak",
+ "storm-creature",
"talking",
"transitions",
+ "truck-ad",
"vfx-clean-plate",
"vfx-day-to-night",
"vfx-water",
"video",
- "video-t2v"
+ "video-t2v",
+ "volcano-ad"
]
},
{
@@ -196,15 +292,31 @@
"gated": true,
"license": "LTX-2.x Community License",
"used_by": [
+ "asmr-host",
+ "asmr-product",
+ "asmr-whisper",
+ "behind-the-scenes",
"camera-copy",
+ "chameleon",
+ "cloud-surf",
+ "idol",
+ "kick-ad",
+ "magic-button",
+ "melting-doodle",
+ "mukbang",
+ "rap-video",
+ "roller-coaster",
"speak",
+ "storm-creature",
"talking",
"transitions",
+ "truck-ad",
"vfx-clean-plate",
"vfx-day-to-night",
"vfx-water",
"video",
- "video-t2v"
+ "video-t2v",
+ "volcano-ad"
]
}
]
diff --git a/tools/template_apps.py b/tools/template_apps.py
index 80db5c2..8dc2e2f 100644
--- a/tools/template_apps.py
+++ b/tools/template_apps.py
@@ -7,11 +7,15 @@
python tools/template_apps.py --lab # prints the lab command that makes every template's example
Each entry below is a Higgsfield app (its name in `feature`) rebuilt as a prompt template on one of
-three bases: FLUX 3 Image Klein's edit pipeline (apps/edit, one reference image; the same workflow
-behind edit, character and product-shot) or Z-Image Turbo (apps/z-image, text to image). The app
-folder gets a copy of the base workflow (built by tools/build_apps.py) and a manifest whose "details"
-input carries the prompt as a bind template; selects feed {placeholders} in that prompt. The base
-app's models and credits are copied as they are.
+these bases, all workflows built by tools/build_apps.py:
+ edit FLUX 3 Image Klein's edit pipeline (apps/edit), one reference image
+ edit2 the same with a second reference image (a garment, another person, a face)
+ z-image Z-Image Turbo (apps/z-image), text to image
+ video LTX-2.5 image-to-video (apps/video), with its own audio
+ qwen-edit Qwen-Image-Edit-2509 (apps/relight) with a different LoRA: Multiple Angles or next-scene
+The app folder gets the base workflow (copied, or with the few edits noted in BASES) and a manifest
+whose "details" input carries the prompt as a bind template; selects feed {placeholders} in that
+prompt. The base app's models and credits are copied as they are (qwen-edit swaps the LoRA).
lab/cases.json gets one "tpl-<id>" showcase case per app (the section after the effects cases is
regenerated in place), on the synthetic references the lab already makes, plus the few new
@@ -27,10 +31,26 @@ from pathlib import Path
ROOT = Path(__file__).resolve().parent.parent
-# the pipelines a template can sit on: node ids for the image, the prompt and the settings
+# the pipelines a template can sit on: node ids for the image(s), the prompt and the settings kept from the base app
BASES = {
- "edit": {"app": "edit", "pack": "image-flux3", "image": ("11", "image"), "prompt": ("4", "text"), "keep": ("steps", "seed"), "output": {"node": "9", "type": "image"}},
- "z-image": {"app": "z-image", "pack": "image", "image": None, "prompt": ("57:27", "text"), "keep": ("width", "height", "steps", "seed"), "output": {"node": "9", "type": "image"}},
+ "edit": {"app": "edit", "section": "image", "pack": "image-flux3", "image": ("11", "image"), "prompt": ("4", "text"), "keep": ("steps", "seed"),
+ "output": {"node": "9", "type": "image"}},
+ "edit2": {"app": "edit", "section": "image", "pack": "image-flux3", "image": ("11", "image"), "image2": ("12", "image"), "prompt": ("4", "text"),
+ "keep": ("steps", "seed"), "output": {"node": "9", "type": "image"}},
+ "z-image": {"app": "z-image", "section": "image", "pack": "image", "image": None, "prompt": ("57:27", "text"), "keep": ("width", "height", "steps", "seed"),
+ "output": {"node": "9", "type": "image"}},
+ "video": {"app": "video", "section": "video", "pack": "video-ltx25", "image": ("395", "image"), "prompt": ("398:376", "value"),
+ "keep": ("duration", "enhance", "aspect", "seed"), "output": {"node": "75", "type": "video"}},
+ "qwen-edit": {"app": "relight", "section": "image", "pack": "image-qwen-edit", "image": ("11", "image"), "prompt": ("15", "value"), "keep": ("seed",),
+ "output": {"node": "12", "type": "image"}},
+}
+
+# LoRAs the qwen-edit templates load instead of Relight (file name as ComfyUI sees it)
+QWEN_LORAS = {
+ "angles": {"repo": "dx8152/Qwen-Edit-2509-Multiple-angles", "file": "镜头转换.safetensors", "gb": 0.24, "license": "Apache-2.0",
+ "credit": "Multiple Angles LoRA by dx8152 (Apache-2.0)"},
+ "next-scene": {"repo": "lovis93/next-scene-qwen-image-lora-2509", "file": "next-scene_lora-v2-3000.safetensors", "gb": 0.3, "license": "MIT",
+ "credit": "next-scene LoRA v2 by lovis93 (MIT)"},
}
# new synthetic references the template cases need (the rest come from the effects section)
@@ -40,7 +60,11 @@ REFS = {
"ref-tpl-sign": ("Photo of a small bakery storefront with a wooden sign above the door that reads OPEN in bold white letters, daylight, photorealistic", 1280, 768,
"Reference with legible text for Edit Text."),
}
+REFS["ref-tpl-garment"] = ("Product photo of a mustard yellow knit cardigan laid flat on a plain white background, top-down, soft even light", 1024, 1024,
+ "A garment on its own, for the two-reference apps.")
PERSON, MAN, WOMAN, PRODUCT, SNEAKER, SCENE, LAKE = "ref-face-a", "ref-fx-man-b", "ref-fx-woman-a", "ref-product", "ref-fx-sneaker", "ref-fx-cafe", "ref-fx-lake"
+WOMAN_B, STREET, STORM, ROAD, GARMENT, CHARACTER = "ref-fx-woman-b", "ref-fx-street", "ref-fx-storm", "ref-fx-road", "ref-tpl-garment", "ref-character"
+PORTRAIT = "9:16 (Portrait Widescreen)"
KEEP_PRODUCT = " Keep the product identical: shape, label, text and colors."
KEEP_FACE = " Keep the face, hairstyle and build identical."
@@ -68,9 +92,9 @@ def txt(id_, label, default="", required=False, help_=None):
T = []
-def A(id_, name, feature, cat, base, image_label, summary, prompt, inputs=(), ref=None, values=None, note=""):
+def A(id_, name, feature, cat, base, image_label, summary, prompt, inputs=(), ref=None, values=None, note="", image2_label=None, ref2=None, lora=None, help_=None):
T.append(dict(id=id_, name=name, feature=feature, cat=cat, base=base, image_label=image_label, summary=summary, prompt=prompt,
- inputs=list(inputs), ref=ref, values=values or {}, note=note))
+ inputs=list(inputs), ref=ref, values=values or {}, note=note, image2_label=image2_label, ref2=ref2, lora=lora, help=help_))
# ---------------------------------------------------------------- ads & products (apps/edit pipeline, a product photo)
@@ -283,6 +307,160 @@ A("color-grade", "Color Grade", "Color Grading", "templates", "edit", "Your phot
("Pastel", "soft pastel"), ("High-contrast B&W", "high-contrast black and white"), ("Golden hour", "golden hour glow")])],
SCENE, {"look": "teal and orange blockbuster"}, "Only the colors change.")
+# ---------------------------------------------------------------- two reference images (edit2: person + garment, scene + person, photo + face)
+A("fashion-factory", "Fashion Factory", "Fashion Factory", "portraits", "edit2", "Your photo",
+ "A fashion catalog photo of you wearing a garment from a second photo.",
+ "A fashion catalog photo: the person from <ref_image_0> wearing the garment from <ref_image_1>, full-length, {scene}. Keep the face identical and the garment's "
+ "cut, color and pattern exactly as in its photo.",
+ [sel("scene", "Scene", [("Studio cyclorama", "in a white studio cyclorama"), ("City street, golden hour", "on a city street at golden hour"), ("Minimalist loft", "in a minimalist loft"),
+ ("Runway", "walking a runway under spotlights")])],
+ WOMAN, {}, "She wears the yellow cardigan; catalog framing.", image2_label="Garment photo", ref2=GARMENT)
+A("try-on", "Virtual Try-On", "AI Stylist", "portraits", "edit2", "Your photo",
+ "Try a garment on your own photo: same pose, same background, new clothes.",
+ "The person from <ref_image_0> wearing the clothing item from <ref_image_1>, same pose, framing and background as <ref_image_0>, natural fit and folds. "
+ "Keep the face and everything else identical; keep the garment's color and pattern.",
+ [], MAN, {}, "Same photo of him, now in the cardigan.", image2_label="Garment photo", ref2=GARMENT)
+A("outfit-swap", "Outfit Swap", "Outfit Swap", "portraits", "edit2", "Your photo",
+ "Wear the whole outfit of the person in another photo.",
+ "The person from <ref_image_0> now dressed in the complete outfit worn by the person in <ref_image_1>, same pose, framing and background as <ref_image_0>. "
+ "Keep the face identical.",
+ [], PERSON, {}, "She wears his red jacket, jeans and sneakers.", image2_label="Photo of the outfit", ref2=CHARACTER)
+A("outfit-shot", "Outfit Shot", "Outfit Shot", "portraits", "edit2", "Model photo",
+ "A studio fashion shot of a garment on a model from your photo.",
+ "A studio fashion photo of the person from <ref_image_0> modeling the garment from <ref_image_1}, {pose}, clean studio backdrop, soft light. "
+ "Keep the face identical and the garment exactly as in its photo.".replace("<ref_image_1}", "<ref_image_1>"),
+ [sel("pose", "Pose", [("Standing", "standing straight, hands relaxed"), ("Walking", "mid-step, walking toward the camera"), ("On a stool", "sitting on a wooden stool")])],
+ WOMAN_B, {}, "Studio shot in the cardigan, same face.", image2_label="Garment photo", ref2=GARMENT)
+A("character-swap", "Character Swap", "Character Swap", "identity", "edit2", "Scene photo",
+ "Replaces the person in a scene with the person from a second photo.",
+ "The scene from <ref_image_0> with its main person replaced by the person from <ref_image_1>: same pose, framing, lighting and clothing as the scene; "
+ "the new person's face, hair and build.",
+ [], SCENE, {}, "The woman sits in the cafe in the man's place and pose.", image2_label="Photo of the new person", ref2=WOMAN)
+A("face-swap", "Face Swap", "Face Swap", "identity", "edit2", "Photo to edit",
+ "Puts the face from a second photo onto the first; for faces you own or have permission to use.",
+ "The photo from <ref_image_0> with the face replaced by the face of the person in <ref_image_1>, matching the original head angle, expression, lighting "
+ "and skin tone; everything else unchanged.",
+ [], PERSON, {}, "Her portrait with the other woman's face; lighting matched.", image2_label="Photo of the face", ref2=WOMAN_B,
+ help_="Only faces you own or have permission to use.")
+
+# ---------------------------------------------------------------- video templates (LTX-2.5 image-to-video, with sound)
+A("behind-the-scenes", "Behind the Scenes", "Behind the Scenes", "clips", "video", "Your photo",
+ "The camera pulls back from your photo to reveal the film set around it.",
+ "The camera pulls back from the photo to reveal a film set around it: crew members, a boom microphone, softbox lights and a clapperboard, people moving about; "
+ "the sound of a busy set, a director calling \"rolling\".",
+ [], PERSON, {"aspect": "16:9 (Widescreen)"}, "A set appears around her as the camera widens.")
+A("asmr-product", "ASMR Product", "ASMR Add-On", "ads", "video", "Product photo",
+ "Your product in an ASMR clip: tapping, scratching, no music.",
+ "An ASMR video: hands with manicured nails slowly tap and scratch the product, turning it under soft light, extreme close-up; crisp tapping and scratching "
+ "sounds, no music, no voice.",
+ [], PRODUCT, {"aspect": PORTRAIT}, "Hands tap the bottle; tapping sounds.")
+A("asmr-whisper", "ASMR Whisper", "ASMR Classic", "clips", "video", "Your photo",
+ "You whispering and tapping in a classic ASMR clip.",
+ "An ASMR video: the person whispers softly toward the camera and taps their fingers on the table, slow gentle movements, warm low light; soft whispering "
+ "and tapping sounds, no music.",
+ [], WOMAN, {"aspect": PORTRAIT}, "Whispering and tapping; calm.")
+A("asmr-host", "ASMR Host", "ASMR Host", "clips", "video", "Your photo",
+ "You hosting an ASMR studio with a big microphone and fairy lights.",
+ "The person sits in a cozy ASMR studio with a large microphone in front of them, fairy lights behind, and whispers into the mic while gently handling a soft "
+ "brush; close whispering sound.",
+ [], MAN, {"aspect": PORTRAIT}, "Studio, mic, whisper.")
+A("chameleon", "Chameleon", "Chameleon", "ads", "video", "Product photo",
+ "A chameleon climbs onto your product and takes its colors.",
+ "A chameleon crawls onto the product and slowly changes its skin to the exact colors and pattern of the product, blinking, macro lens; soft ambient "
+ "jungle sound.",
+ [], SNEAKER, {"aspect": "16:9 (Widescreen)"}, "The chameleon turns red and white like the sneaker.")
+A("magic-button", "Magic Button", "Magic button", "ads", "video", "Product photo",
+ "A glowing button turns the scene around your product magical.",
+ "A hand presses a big glowing button next to the product; the scene transforms into a magical world with floating sparks and soft light, the product "
+ "levitating and turning slowly; a magical chime.",
+ [], PRODUCT, {"aspect": PORTRAIT}, "Button, sparks, the bottle floats.")
+A("truck-ad", "Truck Ad", "Truck Ad", "ads", "video", "Product photo",
+ "Your product printed huge on the side of a passing delivery truck.",
+ "The camera pulls back from the product to reveal it printed huge on the side of a delivery truck driving along a city avenue, the truck passing by with "
+ "sunlight flashing on it; traffic sound.",
+ [], PRODUCT, {"aspect": "16:9 (Widescreen)"}, "The bottle on a truck's side, truck moving.")
+A("kick-ad", "Kick Ad", "Kick Ad", "ads", "video", "Product photo",
+ "Your product kicks the plain grey competition off screen.",
+ "The product stands in the center; plain grey unmarked boxes slide in from the sides and the product knocks them away with a swift cartoon kick, they tumble "
+ "off screen; comedic whoosh and thud sounds.",
+ [], PRODUCT, {"aspect": "16:9 (Widescreen)"}, "Grey boxes tumble; cartoon sounds. No real brands.")
+A("volcano-ad", "Volcano Ad", "Volcano Ad", "ads", "video", "Product photo",
+ "Your product in front of an erupting volcano.",
+ "The product stands on dark volcanic rock in front of an erupting volcano, lava fountains and glowing ash in the background, heat shimmer, slow camera "
+ "push-in; a deep rumbling sound.",
+ [], PRODUCT, {"aspect": "16:9 (Widescreen)"}, "Lava behind the bottle; rumble.")
+A("roller-coaster", "Roller Coaster", "Roller Coaster", "clips", "video", "Your photo",
+ "You on a roller coaster, camera mounted on the car.",
+ "The person rides a roller coaster: the camera is mounted on the car facing them, the track climbs then drops, hair whipping, hands up, laughing with joy; "
+ "clattering track and wind.",
+ [], WOMAN, {"aspect": PORTRAIT}, "Climb, drop, hair flying.")
+A("storm-creature", "Storm Creature", "Storm Creature", "clips", "video", "Your photo",
+ "A creature of clouds and lightning rises over your scene.",
+ "A towering creature made of storm clouds and lightning rises from the clouds over the landscape, lightning flashing inside its body, rain starting; "
+ "thunder and wind.",
+ [], STORM, {"aspect": "16:9 (Widescreen)"}, "A cloud giant with lightning inside.")
+A("cloud-surf", "Cloud Surf", "Cloud Surf", "clips", "video", "Your photo",
+ "You surfing across pink clouds at sunset.",
+ "The person surfs on a surfboard across fluffy pink clouds at sunset, carving through the cloud tops, spray of cloud mist, dynamic tracking shot; wind "
+ "and a cheerful whoosh.",
+ [], MAN, {"aspect": "16:9 (Widescreen)"}, "Surfboard on clouds; same face.")
+A("mukbang", "Mukbang", "Mukbang", "clips", "video", "Your photo",
+ "You at a table full of food, eating for the camera.",
+ "A mukbang video: the person sits at a table loaded with food in front of a microphone, picks up a piece and eats with relish, nodding at the camera; "
+ "crunching and chewing sounds, a short delighted hum.",
+ [], MAN, {"aspect": PORTRAIT}, "Food, mic, eating sounds.")
+A("melting-doodle", "Melting Doodle", "Melting Doodle", "clips", "video", "Your photo",
+ "Cartoon doodles appear over your photo and melt down the frame.",
+ "Colorful cartoon doodles appear drawn over the scene, then begin to melt and drip down the frame like wet paint, dripping and merging; a soft squelching "
+ "sound.",
+ [], STREET, {"aspect": PORTRAIT}, "Doodles drawn, then dripping.")
+A("rap-video", "Rap Video", "Rap God", "clips", "video", "Your photo",
+ "A rap music video of you: low angle, stage lights, smoke, a beat.",
+ "A rap music video: the person performs toward the camera with confident hand gestures, low-angle shot, flashing stage lights and smoke, quick push-ins; "
+ "a punchy hip-hop beat.",
+ [], MAN, {"aspect": PORTRAIT}, "Performance energy; a beat plays.")
+A("idol", "Idol", "Idol", "clips", "video", "Your photo",
+ "Your K-pop idol moment: spotlights, wind, confetti, a pose.",
+ "A pop idol stage moment: the person stands center stage under spotlights, hair blown by a fan, confetti falling, strikes a pose and smiles at the camera; "
+ "crowd cheering and a pop beat.",
+ [], WOMAN, {"aspect": PORTRAIT}, "Stage, confetti, pose.")
+
+# ---------------------------------------------------------------- camera LoRAs on Qwen-Image-Edit-2509 (the relight pipeline, other LoRA)
+MOVES = [("Move forward", "将镜头向前移动 Move the camera forward."), ("Move back", "将镜头向后移动 Move the camera backward."),
+ ("Move left", "将镜头向左移动 Move the camera left."), ("Move right", "将镜头向右移动 Move the camera right."),
+ ("Move up", "将镜头向上移动 Move the camera up."), ("Move down", "将镜头向下移动 Move the camera down."),
+ ("Rotate 45° left", "将镜头向左旋转45度 Rotate the camera 45 degrees to the left."), ("Rotate 45° right", "将镜头向右旋转45度 Rotate the camera 45 degrees to the right."),
+ ("Top-down view", "将镜头转为俯视 Turn the camera to a top-down view."), ("Low angle", "将镜头转为仰视 Turn the camera to a low-angle view looking up.")]
+A("angles", "Angles", "Angles 2.0", "camera", "qwen-edit", "Your photo",
+ "The same scene from another camera position: move, rotate, look down (Multiple Angles LoRA on Qwen-Image-Edit).",
+ "{move}",
+ [sel("move", "Camera move", MOVES, default=MOVES[6][1])],
+ SCENE, {}, "The cafe seen from 45° to the left; same people and table.", lora="angles")
+A("shots", "Shots", "Shots", "camera", "qwen-edit", "Your photo",
+ "Nine framings of one image: wide, close-up, top-down, low angle and rotations.",
+ "{shot}",
+ [sel("shot", "Shot", [("Wide-angle", "将镜头转为广角镜头 Turn the camera to a wide-angle lens."), ("Close-up", "将镜头转为特写镜头 Turn the camera to a close-up."),
+ ("Extreme close-up", "将镜头转为极特写镜头 Turn the camera to an extreme close-up."), ("Top-down", "将镜头转为俯视 Turn the camera to a top-down view."),
+ ("Low angle", "将镜头转为仰视 Turn the camera to a low-angle view looking up."), ("Rotate 45° left", "将镜头向左旋转45度 Rotate the camera 45 degrees to the left."),
+ ("Rotate 45° right", "将镜头向右旋转45度 Rotate the camera 45 degrees to the right."), ("Rotate 90° left", "将镜头向左旋转90度 Rotate the camera 90 degrees to the left."),
+ ("From behind", "将镜头转到人物背后 Move the camera behind the subject.")])],
+ LAKE, {}, "The lake scene as a wide-angle shot.", lora="angles")
+A("zooms", "Zooms", "Zooms", "camera", "qwen-edit", "Your photo",
+ "Zoom levels of one image, from a wide view to an extreme close-up.",
+ "{zoom}",
+ [sel("zoom", "Zoom", [("Much wider", "将镜头大幅向后移动并转为广角镜头 Move the camera far back and turn it to a wide-angle lens."), ("Wider", "将镜头向后移动 Move the camera backward."),
+ ("Closer", "将镜头向前移动 Move the camera forward."), ("Close-up", "将镜头转为特写镜头 Turn the camera to a close-up."),
+ ("Extreme close-up", "将镜头转为极特写镜头 Turn the camera to an extreme close-up.")], default="将镜头转为特写镜头 Turn the camera to a close-up.")],
+ STREET, {}, "A close-up of the woman in the raincoat.", lora="angles")
+A("whats-next", "What's Next", "What's Next?", "camera", "qwen-edit", "Your photo",
+ "The next shot of your scene: 8 cinematic continuations (next-scene LoRA on Qwen-Image-Edit).",
+ "Next Scene: {next}",
+ [sel("next", "Continuation", [("Pull back to reveal", "the camera pulls back to reveal the wider surroundings"), ("Push in to a close-up", "the camera pushes in to a close-up on the main subject"),
+ ("Pan right", "the camera pans right, revealing more of the scene"), ("Someone enters", "a new character walks into the frame from the left"),
+ ("Dusk falls", "time passes and the light turns to dusk, lights coming on"), ("Rain starts", "the weather turns and rain begins to fall"),
+ ("Tilt up to the sky", "the camera tilts up toward the sky"), ("Reverse angle", "the camera cuts to the reverse angle, looking back the other way")])],
+ ROAD, {}, "The next shot of the road: wider, more landscape.", lora="next-scene")
+
def check() -> None:
ids = [t["id"] for t in T]
@@ -301,13 +479,15 @@ def check() -> None:
assert used <= have, (t["id"], used - have)
assert have <= used, (t["id"], have - used)
for i in t["inputs"]:
- assert i["id"] not in ("image", "details", "steps", "seed", "width", "height"), (t["id"], i["id"])
+ assert i["id"] not in ("image", "image2", "details", "steps", "seed", "width", "height", "duration", "enhance", "aspect"), (t["id"], i["id"])
if t["base"] == "z-image":
assert t["ref"] is None and t["image_label"] is None
else:
assert t["ref"], t["id"]
- for k in t["values"]:
- assert k in have, (t["id"], k)
+ assert (t["base"] == "edit2") == bool(t["image2_label"] and t["ref2"]), t["id"]
+ assert (t["base"] == "qwen-edit") == (t["lora"] in QWEN_LORAS), t["id"]
+ for k in t["values"]: # lab values name the template's own inputs or the settings kept from the base app
+ assert k in have or k in BASES[t["base"]]["keep"], (t["id"], k)
def manifest(t: dict) -> dict:
@@ -317,30 +497,64 @@ def manifest(t: dict) -> dict:
inputs = []
if base["image"]:
inputs.append({"id": "image", "type": "image", "label": t["image_label"], "required": True, "limits": {"max_mb": 20},
- "bind": [{"node": base["image"][0], "field": base["image"][1]}]})
+ **({"help": t["help"]} if t["help"] else {}), "bind": [{"node": base["image"][0], "field": base["image"][1]}]})
+ if base.get("image2"):
+ inputs.append({"id": "image2", "type": "image", "label": t["image2_label"], "required": True, "limits": {"max_mb": 20},
+ "bind": [{"node": base["image2"][0], "field": base["image2"][1]}]})
inputs += t["inputs"]
# the default "" matters: an optional input with no value is skipped everywhere, and the template with it
inputs.append({"id": "details", "type": "textarea", "label": "Extra direction", "default": "", "help": "Added at the end of the app's prompt.",
"bind": [{"node": base["prompt"][0], "field": base["prompt"][1], "template": t["prompt"] + " {details}"}]})
inputs += [keep[k] for k in base["keep"]]
test = {"image": "test/portrait.png"} if base["image"] else {}
+ if base.get("image2"):
+ test["image2"] = "test/second.png"
+ if base["section"] == "video":
+ test["duration"] = 2
test.update(t["values"])
+ models, credits = src["models"], list(src.get("credits", []))
+ notes = f"Template app written by tools/template_apps.py: the workflow is apps/{base['app']}'s (built by tools/build_apps.py)"
+ if t["base"] == "edit2":
+ notes += " plus a second LoadImage on ReferenceMultiLatent's images.image1"
+ if t["lora"]:
+ lora = QWEN_LORAS[t["lora"]]
+ models = [m for m in models if "Relight" not in m["file"]] + [{"repo": lora["repo"], "file": lora["file"], "dir": "loras", "gb": lora["gb"], "phase": "denoise", "license": lora["license"]}]
+ credits = [c for c in credits if "Relight" not in c]
+ credits.insert(1, lora["credit"])
+ notes += f" with {lora['file']} in place of the Relight LoRA and no Relight trigger"
+ notes += "; the prompt below is fixed and \"Extra direction\" is appended to it."
return {
"$schema": "../../schema/app.schema.json",
- "id": t["id"], "name": t["name"], "section": "image",
+ "id": t["id"], "name": t["name"], "section": base["section"],
"summary": t["summary"],
"higgsfield_feature": t["feature"],
"pack": base["pack"], "status": "ready", "workflow": "workflow.api.json",
- "source": {**src.get("source", {}), "notes": f"Template app written by tools/template_apps.py: the workflow is apps/{base['app']}'s "
- f"(built by tools/build_apps.py) and the prompt below is fixed; \"Extra direction\" is appended to it."},
+ "source": {**src.get("source", {}), "notes": notes},
"inputs": inputs,
"outputs": [{**base["output"], "label": "Result"}],
- "models": src["models"],
- "credits": src.get("credits", []),
+ "models": models,
+ "credits": credits,
"test": test,
}
+def workflow(t: dict) -> dict:
+ """The base app's workflow, with the few edits a base needs."""
+ base = BASES[t["base"]]
+ api = json.loads((ROOT / "apps" / base["app"] / "workflow.api.json").read_text(encoding="utf-8"))
+ if t["base"] == "edit2": # a second reference image into the same ReferenceMultiLatent
+ api["12"] = {"class_type": "LoadImage", "inputs": {"image": "second.png"}, "_meta": {"title": "Second reference"}}
+ api["17"]["inputs"]["images.image1"] = ["12", 0]
+ if t["lora"]: # the relight pipeline with another LoRA; node 14 prepends the Relight trigger, so it is blanked
+ api["10:118"]["inputs"]["lora_name"] = QWEN_LORAS[t["lora"]]["file"]
+ api["14"]["inputs"]["string_a"] = ""
+ for n in api.values():
+ if n["class_type"] == "SaveImage" or n["class_type"] == "SaveVideo" or n["class_type"] == "CreateVideo":
+ if "filename_prefix" in n["inputs"]:
+ n["inputs"]["filename_prefix"] = f"freefield/{t['id']}"
+ return api
+
+
def cases() -> list[str]:
j = lambda o: json.dumps(o, ensure_ascii=False) # noqa: E731
lines = []
@@ -348,7 +562,7 @@ def cases() -> list[str]:
lines.append(f' {{"id": "{rid}", "app": "z-image", "values": {j({"prompt": prompt, "width": w, "height": h})},\n "note": {j(note)}}},')
lines.append("")
for t in T:
- vals = ({"image": f"@case:{t['ref']}"} if t["ref"] else {}) | t["values"]
+ vals = ({"image": f"@case:{t['ref']}"} if t["ref"] else {}) | ({"image2": f"@case:{t['ref2']}"} if t["ref2"] else {}) | t["values"]
note = j(f"{t['feature']}. {t['note']}")
lines.append(f' {{"id": "tpl-{t["id"]}", "showcase": true, "app": "{t["id"]}", "values": {j(vals)},\n'
f' "note": {note}}},')
@@ -377,8 +591,9 @@ def write_cases() -> int:
def readme() -> str:
rows = ["| App | Rebuilds | Base | Inputs |", "|---|---|---|---|"]
for t in T:
- extra = ", ".join(i["label"].lower() for i in t["inputs"])
- rows.append(f"| `{t['id']}` | {t['feature']} | `{BASES[t['base']]['app']}` | {t['image_label'].lower() + (', ' if extra else '') if t['image_label'] else ''}{extra} |")
+ ins = [x.lower() for x in (t["image_label"], t["image2_label"]) if x] + [i["label"].lower() for i in t["inputs"]]
+ base = BASES[t["base"]]["app"] + (f" + {QWEN_LORAS[t['lora']]['repo'].split('/')[1]}" if t["lora"] else "")
+ rows.append(f"| `{t['id']}` | {t['feature']} | `{base}` | {', '.join(ins)} |")
return "\n".join(rows)
@@ -402,8 +617,7 @@ def main(argv=None) -> int:
d = ROOT / "apps" / t["id"]
d.mkdir(parents=True, exist_ok=True)
(d / "app.json").write_text(json.dumps(manifest(t), indent=2, ensure_ascii=False) + "\n", encoding="utf-8")
- wf = (ROOT / "apps" / BASES[t["base"]]["app"] / "workflow.api.json").read_text(encoding="utf-8")
- (d / "workflow.api.json").write_text(wf, encoding="utf-8")
+ (d / "workflow.api.json").write_text(json.dumps(workflow(t), indent=2, ensure_ascii=False) + "\n", encoding="utf-8")
print(f"apps/: {len(T)} template apps written")
print(f"lab/cases.json: {write_cases()} cases")
return 0
diff --git a/web/js/catalog.js b/web/js/catalog.js
index aa3ac62..43ba206 100644
--- a/web/js/catalog.js
+++ b/web/js/catalog.js
@@ -21,12 +21,14 @@ export async function loadCatalog() {
/** Apps grouped like a store: a card can only sit in one row. */
export const CATEGORIES = [
{ id: "create", apps: ["z-image", "cinema-studio", "styles", "video-t2v", "video", "music", "sfx"] },
- { id: "identity", apps: ["character", "speak", "talking", "recast", "motion-transfer"] },
- { id: "camera", apps: ["camera-moves", "camera-copy", "reshoot", "transitions"] },
+ { id: "identity", apps: ["character", "speak", "talking", "recast", "motion-transfer", "character-swap", "face-swap"] },
+ { id: "camera", apps: ["camera-moves", "camera-copy", "reshoot", "transitions", "angles", "shots", "zooms", "whats-next"] },
{ id: "style", apps: ["relight", "style-reference", "upscale", "upscale-video", "smooth", "remove-bg"] },
{ id: "vfx", apps: ["effects", "edit", "vfx-clean-plate", "vfx-day-to-night", "vfx-water"] },
- { id: "ads", apps: ["product-shot", "poster", "packshot", "macro-scene", "macro-product", "giant-product", "billboard", "graffiti-ad", "fridge-ad", "ugc-shot"] },
- { id: "portraits", apps: ["headshot", "commercial-faces", "style-snap", "skin-enhancer", "outfit-vending", "glitter-sticker"] },
+ { id: "ads", apps: ["product-shot", "poster", "packshot", "macro-scene", "macro-product", "giant-product", "billboard", "graffiti-ad", "fridge-ad", "ugc-shot",
+ "asmr-product", "chameleon", "magic-button", "truck-ad", "kick-ad", "volcano-ad"] },
+ { id: "portraits", apps: ["headshot", "commercial-faces", "style-snap", "skin-enhancer", "outfit-vending", "glitter-sticker", "fashion-factory", "try-on", "outfit-swap", "outfit-shot"] },
+ { id: "clips", apps: ["behind-the-scenes", "asmr-whisper", "asmr-host", "roller-coaster", "storm-creature", "cloud-surf", "mukbang", "melting-doodle", "rap-video", "idol"] },
{ id: "play", apps: ["game-styles", "plushie", "render-3d", "figure-3d", "pixel-game", "brick-toy", "sketch-to-real", "paint-app", "mural", "micro-beasts", "meme"] },
{ id: "templates", apps: ["j-magazine", "cafe-60s", "victory-card", "j-poster", "mascot", "burning-sunset", "comic-book", "giallo", "renaissance", "japanese-show", "breakdown", "edit-text", "image-effects", "color-grade"] },
];
@@ -39,6 +41,8 @@ export const VIDEO_TABS = [
{ id: "create", apps: ["video", "effects", "video-t2v", "speak", "talking", "transitions", "camera-moves"] },
{ id: "edit", apps: ["recast", "vfx-clean-plate", "vfx-day-to-night", "vfx-water", "upscale-video", "smooth", "reshoot"] },
{ id: "motion", apps: ["motion-transfer", "camera-copy"] },
+ { id: "templates", apps: ["behind-the-scenes", "asmr-product", "asmr-whisper", "asmr-host", "chameleon", "magic-button", "truck-ad", "kick-ad", "volcano-ad",
+ "roller-coaster", "storm-creature", "cloud-surf", "mukbang", "melting-doodle", "rap-video", "idol"] },
];
export const videoTabOf = (id) => VIDEO_TABS.find((tab) => tab.apps.includes(id))?.id ?? null;
diff --git a/web/js/i18n.js b/web/js/i18n.js
index cedfac3..cd6f7db 100644
--- a/web/js/i18n.js
+++ b/web/js/i18n.js
@@ -41,8 +41,9 @@ const en = {
"cat.portraits.t": "Portraits & looks", "cat.portraits.d": "Headshots, campaigns and outfits from one selfie.",
"cat.play.t": "Toys & games", "cat.play.d": "Your photo as a plushie, a figure, pixel art, a meme.",
"cat.templates.t": "Templates", "cat.templates.d": "One-click looks: magazine covers, paintings, comics, color grades.",
+ "cat.clips.t": "Video templates", "cat.clips.d": "Your photo in a ready-made clip: ASMR, roller coaster, idol stage, cloud surfing.",
"cat.more.t": "More", "cat.more.d": "",
- "vtab.create": "Create video", "vtab.edit": "Edit video", "vtab.motion": "Motion control",
+ "vtab.create": "Create video", "vtab.edit": "Edit video", "vtab.motion": "Motion control", "vtab.templates": "Templates",
"ws.history": "History", "ws.library": "Library", "ws.how": "How it works",
"ws.create.title": "Start creating with", "ws.prompt.ph": "Describe the scene you imagine", "ws.prompt.opt": "{label} (optional): the app's prompt is already set", "ws.add": "Add {what}",
"ws.generate": "Generate", "ws.est": "GPU time of the lab example ({gpu})", "ws.count": "Generations",
@@ -119,8 +120,9 @@ const es = {
"cat.portraits.t": "Retratos y looks", "cat.portraits.d": "Headshots, campañas y outfits desde una selfie.",
"cat.play.t": "Juguetes y juegos", "cat.play.d": "Tu foto como peluche, figura, pixel art o meme.",
"cat.templates.t": "Plantillas", "cat.templates.d": "Looks de un clic: portadas, pinturas, cómics, color grading.",
+ "cat.clips.t": "Plantillas de video", "cat.clips.d": "Tu foto en un clip listo: ASMR, montaña rusa, escenario de idol, surf en nubes.",
"cat.more.t": "Más", "cat.more.d": "",
- "vtab.create": "Crear video", "vtab.edit": "Editar video", "vtab.motion": "Control de movimiento",
+ "vtab.create": "Crear video", "vtab.edit": "Editar video", "vtab.motion": "Control de movimiento", "vtab.templates": "Plantillas",
"ws.history": "Historial", "ws.library": "Biblioteca", "ws.how": "Cómo funciona",
"ws.create.title": "Empieza a crear con", "ws.prompt.ph": "Describe la escena que imaginas", "ws.prompt.opt": "{label} (opcional): el prompt de la app ya está puesto", "ws.add": "Agregar {what}",
"ws.generate": "Generar", "ws.est": "Tiempo de GPU del ejemplo del lab ({gpu})", "ws.count": "Generaciones",