Text-to-Image
Diffusers
English
image-editing
SVDQuant
Qwen-Image-Edit-2509
Diffusion
Quantization
ICLR2025
Instructions to use nunchaku-ai/nunchaku-qwen-image-edit-2509 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use nunchaku-ai/nunchaku-qwen-image-edit-2509 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("nunchaku-ai/nunchaku-qwen-image-edit-2509", torch_dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
with lightining, anime to realistic style transfer doesnt work.
#4
by lalisa-2424 - opened
I dont know if its me doing something wrong but with lightining, anime to realistic style transfer doesnt work. It just gives the same image without any changes or slight irrelevant changes. I tried svdq-fp4_r32-qwen-image-edit-2509-lightningv2.0-4steps.safetensors, and svdq-fp4_r128-qwen-image-edit-2509-lightningv2.0-8steps.safetensors. with different prompts, different images and different adjustments to be sure. I have 5060ti. It works with the svdq-fp4_r32-qwen-image-edit-2509.safetensors btw.
It slighly better with some images when I adjust CFG to 4.0 with the lighting version but naturally it comes with artifacts, unwanted changes and loss of quality.