Instructions to use showlab/OmniConsistency with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use showlab/OmniConsistency 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("showlab/OmniConsistency", 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
Compatibility with Invoke, and Usage Question
Hi!
I'm a bit of a noob with this.
Does OmniConsistency.safetensor work with Invoke Community edition? I get this error when trying to install: "Model install error No valid config found"
Maybe it has something to do with flux pipeline?
Also can I use a reference image for the style or I need to make my own LoRA?
Thanks!
Hi, @Dylanyz
Thanks for your interest!
I haven’t tried Invoke Community edition myself, so I’m not sure whether it works with our method. However, we’ve just released a ComfyUI node for OmniConsistency:
🔗 https://github.com/lc03lc/Comfyui_OmniConsistency
You can give it a try within ComfyUI.
As for using a reference image for style—this isn’t supported yet. We’re currently working on integrating IP-Adapter to enable that functionality. For now, you’ll need to either train your own LoRA or use existing community LoRAs based on Flux.1.