File size: 3,566 Bytes
4d4ea09 5813c59 4d4ea09 39c69b7 9245fa1 39c69b7 4d4ea09 39c69b7 9245fa1 4d4ea09 9245fa1 4d4ea09 9245fa1 4d4ea09 9245fa1 4d4ea09 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 |
# LTX-2 Image-to-Video Adapter LoRA
A high-rank LoRA adapter for [LTX-Video 2](https://github.com/Lightricks/LTX-Video) that substantially improves image-to-video generation quality. No complex workflows, no image preprocessing, no compression tricks -- just a direct image embedding pipeline that works.
## What This Is
Out of the box, getting LTX-2 to reliably infer motion from a single image requires heavy workflow engineering -- ControlNet stacking, image preprocessing, latent manipulation, and careful node routing. The purpose of this LoRA is to eliminate that complexity entirely. It teaches the model to produce solid image-to-video results from a straightforward image embedding, no elaborate pipelines needed.
Trained on **30,000 generated videos** spanning a wide range of subjects, styles, and motion types, the result is a highly generalized adapter that strengthens LTX-2's image-to-video capabilities without any of the typical workflow overhead.
### Key Specs
| Parameter | Value |
|-----------|-------|
| **Base Model** | LTX-Video 2 |
| **LoRA Rank** | 256 |
| **Training Set** | ~30,000 generated videos |
| **Training Scope** | Visual only (no explicit audio training) |
## What It Does
- **Improved image fidelity** -- the generated video maintains stronger adherence to the source image with less drift or distortion across frames.
- **Better motion coherence** -- subjects move more naturally and consistently throughout the clip.
- **Broader generalization** -- performs well across diverse subjects and scenes without needing per-category tuning.
### A Note on Audio
Audio was **not** explicitly trained into this LoRA. However, due to the nature of how LTX-2 handles its latent space, there are subtle shifts in audio output compared to the base model. This is a side effect of the training process, not an intentional feature.
## Usage (ComfyUI)
1. Place the LoRA file in your `ComfyUI/models/loras/` directory.
2. Add an **LTX-2** model loader node and load the base LTX-2 checkpoint.
3. Add a **Load LoRA** node and select this adapter.
4. Connect an **image embedding** node with your source image.
5. Add your text prompt and generate.
### Workflow
## Examples
Reference videos demonstrating the adapter's output quality:
<video src="https://huggingface.co/MachineDelusions/LTX-2_Image2Video_Adapter_LoRa/resolve/main/assets/AnimateDiff_00740.mp4" autoplay loop muted playsinline></video>
<video src="https://huggingface.co/MachineDelusions/LTX-2_Image2Video_Adapter_LoRa/resolve/main/assets/AnimateDiff_00774.mp4" autoplay loop muted playsinline></video>
<video src="https://huggingface.co/MachineDelusions/LTX-2_Image2Video_Adapter_LoRa/resolve/main/assets/AnimateDiff_00777.mp4" autoplay loop muted playsinline></video>
<video src="https://huggingface.co/MachineDelusions/LTX-2_Image2Video_Adapter_LoRa/resolve/main/assets/AnimateDiff_00778.mp4" autoplay loop muted playsinline></video>
## Model Details
- **Architecture:** LoRA (Low-Rank Adaptation) applied to LTX-Video 2's transformer layers
- **Rank 256** provides a high-capacity adaptation while remaining efficient to load and merge
- **Training data** was intentionally diverse to avoid overfitting to any single domain, producing a general-purpose image-to-video adapter rather than a style-specific fine-tune
## License
Please refer to the [LTX-Video license](https://github.com/Lightricks/LTX-Video) for base model terms.
|