Instructions to use segmind/SSD-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use segmind/SSD-1B with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("segmind/SSD-1B", 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
stable-diffusion.cpp
please note that SSD-1B is supported by stable-diffusion.cpp for some time.
https://github.com/leejet/stable-diffusion.cpp
Needless to say, but SSD-1B used together with sd.cpp ( https://github.com/leejet/stable-diffusion.cpp ) is 1st choice
for exeuting text-to-image jobs with a feeling of SDXL on low powered Android smartphones without using python.
You could do this inside termux environment, but even without termux,
for example using this app: https://github.com/akleine/sdcpp-on-android
Note: unlike other tiny UNet-SD1x-models this SSD-1B does not require any conversion steps, it is just ready for run,
see also https://github.com/leejet/stable-diffusion.cpp/blob/master/docs/distilled_sd.md