Instructions to use SG161222/Realistic_Vision_V3.0_VAE with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use SG161222/Realistic_Vision_V3.0_VAE with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("SG161222/Realistic_Vision_V3.0_VAE", 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
Did you encounter issues when converting the weights from safetensors to this format?
#9
by jerry23 - opened
First thank you for providing such a nice model!
I've tried converting your model from safetensors to diffusers's format using their from_ckpt (before you release this), but I encountered something like this: https://github.com/huggingface/diffusers/issues/3595
I still can finish the conversion, but I wonder if that will cause negative effect on the output quality, so I'm curious if you encountered similar issues?