Text-to-Image
Diffusers
Safetensors
StableDiffusionPipeline
Base Model
Everything
runwayml
stable-diffusion
stable-diffusion-diffusers
Instructions to use Yntec/stable-diffusion-v1-5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Yntec/stable-diffusion-v1-5 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Yntec/stable-diffusion-v1-5", 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
V-prediction or Epsilon
#2
by Crowyote - opened
When downloading from Invoke's GUI repository connection it asks me which to prefer. I'm guessing v-prediction, but clarifying this would help.
This model uses epsilon:
https://huggingface.co/Yntec/StableDiffusion768/blob/main/scheduler/scheduler_config.json
v-prediction is recommended for Stable Diffusion 2, 2.1 and SDXL based models.
Do you think Epsilon would be required or just forget it?
Edit: I see you mentioned "epsilon."
I selected nothing and it seems ok. Maybe that's the default on invoke.
Crowyote changed discussion status to closed
v-prediction was implemented on the 2.0 and beyond era, I don't think people tested it to see how Stable Diffusion 1.5 based models would perform with it.