import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("LiconStudio/LTX-2.5-Multiple-Subject-Reference", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Licon MSR V1 for LTX-2.5
Overview
Licon MSR V1 is a multi-reference LoRA trained for LTX-2.5.
It uses the Multiple Subject Reference (MSR) approach to encode multiple reference images as visual tokens in the same latent space as the target video. Each reference is assigned a learned slot embedding and a distinct negative temporal position, allowing target video tokens to retrieve character, clothing, object, and scene information through the model's native self-attention layers.
Key Features
- Supports up to five reference images
- Preserves multiple characters, clothing, objects, and backgrounds
- Learned slot embeddings distinguish different references
- Native self-attention retrieval of reference details
- Supports multi-subject and subject-object composition
- Designed specifically for the LTX-2.5 architecture
Usage
ComfyUI inference requires ComfyUI-LTX2.5-MSR. A sample workflow is included in the plugin repository.
Usage Tips
- Describe each reference image clearly in the prompt.
- Use consistent labels such as
Image 1,Image 2, andImage 3. - Clearly specify subject actions and spatial relationships.
- Specify which reference provides the character, object, clothing, or background.
Examples
| Example 03 | Example 06 | Example 07 |
|---|---|---|
|
Reference Images
|
Reference Images
|
Reference Images
|
- Downloads last month
- 641
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for LiconStudio/LTX-2.5-Multiple-Subject-Reference
Base model
Lightricks/LTX-2.5