import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-Kontext-dev", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("George0667/Kontext_Removal_Lora")
prompt = "remove the object in the red frames"
image = pipe(prompt).images[0]Kontext_Removal_Lora
Model description
Intro:
This is a model for removing objects in images, trained on black-forest-labs/FLUX.1-Kontext-dev. The format is suitable for ComfyUI and is also provided with Workflow.
Usage:
- Loading the workflow to ComfyUI
- Upload the input images and draw a mask for input.
- Running and got the results
NOTE: A custom node for drawing a red box should be used for image preprocessing.
Refer to: https://github.com/George0726/ComfyUI-video-accessory
Examples:
Contact:
If you have further questions, don't hesitate to email me.
Trigger words
You should use remove the object in the red frames to trigger the image generation.
Download model
Download them in the Files & versions tab.
- Downloads last month
- 9
Model tree for George0667/Kontext_Removal_Lora
Base model
black-forest-labs/FLUX.1-Kontext-dev