Instructions to use Lakonik/pi-Qwen-Image with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Lakonik/pi-Qwen-Image with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Lakonik/pi-Qwen-Image", 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
Amazing speed with great quality with ComfyUI native models.
#2
by vladulidlo - opened
I want to thank you for releasing this!
I was using gguf, then nunchaku-int4 with significant color shift and there were obvious quality/speed issues. The fact that with your project we can now use FP8 + Loras without the need of conversion and with significant speed advantage and adjustable low steps without sacrificing quality is amazing.
Compared to nunchaku, your project is much easier to use and manage, faster first run and consequent runs.
I'm hoping and looking forward for pi-Qwen-Image-Edit-2509 release!