Instructions to use dataautogpt3/ProteusV0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use dataautogpt3/ProteusV0.1 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("dataautogpt3/ProteusV0.1", dtype=torch.bfloat16, device_map="cuda") prompt = "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Errors, and excessive RAM overload
#1
by tintwotin - opened
I'll have to uncomment these lines to get it working:
use_safetensors=True,
variant="fp16"
When loading the model the RAM shoot-up using some 28 GB. and then crashes. I have never seen this behavior with any other image model. Normally it's the VRAM on the GPU filling up?
wow! i'll do some testing and change the example if it keeps having this issue! thank you so much for the feedback nonetheless.
I believe its just because there isn't a fp16 safetensors version uploaded yet.
I believe this issue has been resolved.
dataautogpt3 changed discussion status to closed
I believe he uploaded a fp16 safetensors version, but it wasn't the diffusers version.