Instructions to use prithivMLmods/QIE-2511-FlatLog2TrueColor with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use prithivMLmods/QIE-2511-FlatLog2TrueColor 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("Qwen/Qwen-Image-Edit-2511", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("prithivMLmods/QIE-2511-FlatLog2TrueColor") 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] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
What is the difference between FlatLog2TrueColor and BW2TrueColor?
https://huggingface.co/prithivMLmods/QIE-2511-FlatLog2TrueColor
https://huggingface.co/prithivMLmods/QIE-2511-BW2TrueColor
What is the difference between these two LoRAs? They both seem to convert black‑and‑white images into color images.
Hey, @makisekurisu-jp
Logarithmic Gamma Curve (flat) and Black & White images are not the same!
Log images are desaturated and low in contrast because the dynamic range compresses both highlights and shadows. Because of this, they appear flat and usually require color grading, which makes post-processing easier. In contrast, Black & White images have high contrast and contain no color information (monochrome).
In simple terms: B&W images are grayscale (no color) & Log images are technically full color, but they appear muted and flat.
So both adapters are trained differently, based on the respective logit distributions of their data.