Instructions to use madebyollin/taesd with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use madebyollin/taesd with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("madebyollin/taesd", 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
More Clarification Required on the Distillation Scheme
#7
by TalhaAhmed2000 - opened
Hello. Hope you are doing well. I have found your work on the distillation of SD-VAE quite practical and simple to use. However, I was wondering if you could share the details of the 'distillation' scheme of madebyollin/taesd model (which has ~1.22 M parameters for encoder & decoder each). Please guide us to any doc, website (if there is any) which we can use to better understand the procedure.
Thank you for your time & consideration.
See the discussions on GitHub, e.g.
- https://github.com/madebyollin/taesd/issues/11
- https://github.com/madebyollin/taesd/issues/16
- https://github.com/madebyollin/taesd/issues/19
- https://github.com/madebyollin/taesd/issues/21
- https://github.com/madebyollin/taesd/issues/27
There are also other similar projects like https://github.com/bytedance/Hybrid-SD which may be useful reference