Instructions to use ByteDance/Hyper-SD with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use ByteDance/Hyper-SD with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("ByteDance/Hyper-SD") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Feedback
#31
by YaTharThShaRma999 - opened
After testing all the models, (all steps, and cfg 8 step), I have some feedback.
The 1 step version is spectacular, sdxl lightning and turbo step don’t come close. It’s very close quality to base sdxl.
Higher steps are great as well but I would say I prefer sdxl lightning 4 step over 2 and 4 step hyper sdxl. The 8 step is similar quality.
The cfg 8 step one can perform well sometimes but sometimes messes up often as well.
Overall great work, I would not have imagined ever that a model could generate such good images in just a single step!
renyuxi changed discussion status to closed