Image-to-Image
Diffusers
Safetensors
English
Image-to-Image
ControlNet
Diffusers
QwenImageControlNetInpaintPipeline
Qwen-Image
Instructions to use InstantX/Qwen-Image-ControlNet-Inpainting with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use InstantX/Qwen-Image-ControlNet-Inpainting with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("InstantX/Qwen-Image-ControlNet-Inpainting", dtype=torch.bfloat16, device_map="cuda") prompt = "Turn this cat into a dog" input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cat.png") image = pipe(image=input_image, prompt=prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Training details
#11
by qingzwang - opened
Hi authors:
The README shows that the dataset is composed of 10M images and batch size is 128, training steps is 65k, so the model is trained around one epoch, right?
Recently, I am trying to reproduce this model, but I only have 100k images and I trained the model 100 epochs, but it failed to generate satisfying images.
Its fire that you are even tried at it!!!
Its fire that you are even tried atlq
Its fire that you are even tried at it!!!
Finally, I give up and just distill it to a few-step generator.