README.md

---
tags:
- lora
- diffusers
- template:diffusion-lora
- text-to-video
- image-to-video
- video-to-video
- image-to-image
- lightx2v
base_model: ByteDance/Bernini-R
instance_prompt: null
license: apache-2.0
---
# Bernini-R LightX2V LoRAs (4-step)


<video controls width="100%" height="480">
  <source src="https://huggingface.co/rzgar/Bernini-R-LightX2V-4step-loras/resolve/main/Workflow/ComfyUI_00032.mp4" type="video/mp4">
  Your browser does not support the video tag.
</video>
## Model description 

## Why not lightx2v_T2V_14B_cfg_step_distill_v2_lora_rank64_bf16?

The ComfyUI Bernini-R workflow uses a single **Wan 2.1** T2V lightx2v LoRA, reused at **3.0** (high) and **1.5** (low).

In practice that setup often shows:

- Blurry fingers on movement  
- Laggy or glitchy motion when things move quickly  

These LightX2V LoRAs behave like the Wan 2.2 4-step stack , clean motion, sharp hands , at **1.0 strength** on both stages.

## Files

| File | Use on |
|------|--------|
| &#x60;Bernini-R_LightX2V_high_noise.safetensors&#x60; | Bernini-R **high noise** |
| &#x60;Bernini-R_LightX2V_low_noise.safetensors&#x60; | Bernini-R **low noise** |

**Strength: 1.0** on both (same as the official Wan 2.2 4-step workflow).

**Steps: 4**

**Sampler**: dpmpp_2m_sde

**Scheduler**: sgm_uniform


## Credits
[**LightX2V**](https:&#x2F;&#x2F;huggingface.co&#x2F;lightx2v)


## Download model


[Download](/rzgar/Bernini-R-LightX2V-4step-loras/tree/main) them in the Files & versions tab.