Instructions to use stabilityai/stable-diffusion-3.5-large with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use stabilityai/stable-diffusion-3.5-large with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-3.5-large", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Any way to avoid downloading the whole model again? Only transformer updated!
#14
by DealayLomoi - opened
Any way to avoid downloading the whole model again? Only transformer updated! VAE, t5, clip they all the same... They already exsit in my sd3-medium cache...
According to HF Hub Docs, you can use huggingface-cli command to download a single file from a certain repo.
For your situation, if you want to download sd3.5_large.safetensors only, you can use the following command:
huggingface-cli download stabilityai/stable-diffusion-3.5-large sd3.5_large.safetensors --local-dir . --token <YOUR_ACCESS_TOKEN>
The downloaded file will be placed in your current directory.
thank u martin!