import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("lynaNSFW/LTX2BFN", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("lynaNSFW/LTX2.3_NSFW_motion")
prompt = "-"
image = pipe(prompt).images[0]LTX2.3_NSFW motion
.png)
- Prompt
- -
Model description
Updated version for LTX 2.3: Works on 2d 3d realistic, furry non furry etc. I use it at a strength of 0.7 but you might need to adjust the strength depending on your workflow and lora stack. I use with no distilled lora and I use the distilled model. I highly recommend using wan2gp instead of comfyui. I have gotten much better results with it than I have with comfy plus its not confusing like comfy.
Download model
Download them in the Files & versions tab.
- Downloads last month
- 8,308