Loop ForgePortrait, hair in the breeze
Works, from LTX 2.5 Creative Lab LoRAs, tested locally
- LoRA
- Cinemagraph at 1.5
- Output
- 768×1024, 121 frames at 24fps
- Sampling
- 8 steps, euler_ancestral
- Seed
- 42
- Generated in
- 6 min
- Peak VRAM
- 11.5 GiB
- No LoRA run
- Same graph without the LoRA, seed 42, 6 min
- Ran on
- RTX 3060, 12GB (local)
What was wired in
Only a few strands of hair move and the head stays still. I could not get this without the LoRA, whatever I changed in the prompt.
CINEMAGRAPH_MOTION, tripod locked-off static camera, zero camera movement, the woman is frozen, her face, eyes, smile, head, knitted sweater, the bookshelves and the window remain completely frozen, only a few loose strands of her wavy brown hair sway gently in a soft breeze, everything else stays perfectly still, seamless natural loop.
CINEMAGRAPH_MOTION, tripod locked-off static camera, zero camera movement, only the loose strands of her wavy brown hair sway gently in a soft breeze, her face, eyes, smile, knitted sweater and the bookshelves and window behind her stay perfectly still, seamless natural loop
camera movement, pan, tilt, zoom, parallax, whole image moving, background sliding, person moving, face moving, flicker on entire image, noisy texture, crawling texture, distorted, blurry, low quality
Notes
Writing the prompt in the card's order, with what is frozen first and then what moves, made the face slightly stiller than my first prompt did. The trigger is CINEMAGRAPH_MOTION, at strength 1.5.
Run it yourself
Download everything for this test (.zip, 7.2 MB). It holds the workflow, every input it loads and a README. Copy the inputs into ComfyUI/input, put the models below in place and drag the workflow onto ComfyUI.
| File | Put it in | Size |
|---|---|---|
| ltx-2.5-22b-distilled-transformer-comfy-int8-convrot.safetensors gated | models/diffusion_models | 21.5 GB |
| gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors gated | models/text_encoders | 15.4 GB |
| ltx-2.5-video-vae-bf16.safetensors gated | models/vae | 1.5 GB |
| ltx-2.5-audio-vae-bf16.safetensors gated | models/vae | 364.9 MB |
| ltx-2.5-22b-lora-cinemagraph-0.9.safetensors gated | models/loras | 201.5 MB |
The Lightricks repositories (the LTX 2.5 base model and the LoRAs) are gated. Log in to Hugging Face and accept the terms on each repository page before downloading.
No custom nodes. Every node is part of ComfyUI 0.37.0 (commit 0f74f7fb, frontend 1.53.6).