Instructions to use HighCWu/FLUX.1-dev-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use HighCWu/FLUX.1-dev-4bit with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("HighCWu/FLUX.1-dev-4bit", 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
Questions about LoRA
#5
by tungdop2 - opened
- Can i load normal bf16 LoRA weights to this quantized model?
- Which code did you use to train LoRA?
I think if you were to load a lora you would have to do it in model.py before its quantized. I messed with it a bit and decided just touse flux as a refiner for SDXL.