Instructions to use spooknik/Fluxmania-SVDQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use spooknik/Fluxmania-SVDQ with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("spooknik/Fluxmania-SVDQ", 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
How to Load SVDQ Models
#4
by makisekurisu-jp - opened
makisekurisu-jp changed discussion status to closed
makisekurisu-jp changed discussion status to open
Hello, yes that is correct you need to load them using the Nunchaku FLUX DiT Loader.
Here is a workflow for reference.
Could you help add SVDQ quantization support for the SRPO and OneReward models? Thanks~
Hello again,
Thanks for the suggestion, those are on my to-do list. I have tried to make an SVDQuant of them, but they need a bit of special attention to work correctly. I will give it another look in December.
Hello again,
Thanks for the suggestion, those are on my to-do list. I have tried to make an SVDQuant of them, but they need a bit of special attention to work correctly. I will give it another look in December.
Thank you very much.
